JP6568557B2 - Humanization of rabbit antibodies using a versatile antibody framework - Google Patents

Description

(Related information)
This application claims priority to US 61 / 075,697, filed June 25, 2008, and further US 61 / 155,041 and February 24, 2009, filed February 24, 2009. Claims priority to US 61 / 155,105.

(Background of the Invention)
Monoclonal antibodies, their conjugates and derivatives are of great commercial importance as therapeutic and diagnostic agents. Non-human antibodies usually elicit a strong immune response in patients after a single low dose injection (Schroff, 1985, Cancer Res, 45, 879-85, Shawler, J Immunol, 1985, 135). 1530-5, Dillman, Cancer Biother, 1994, 9, 17-28). Thus, several methods have been developed to reduce the immunogenicity of murine and other rodent antibodies, as well as techniques for generating fully human antibodies using, for example, transgenic mice or phage display. . Chimeric antibodies (eg, Boulianne Nature, 1984, 312, 643-6) that combine rodent variable regions with human constant regions have been genetically engineered to significantly reduce immunogenicity problems ( For example, LoBuglio, Proc Natl Acad Sci, 1989, 86, 4220-4; Clark, Immunol Today, 2000, 21, 397-402). Humanized antibodies have also been constructed. In a humanized antibody, the rodent sequence of the variable region itself has been genetically engineered to be as close to the human sequence as possible while preserving at least the original CDR, or the CDR of the rodent antibody is a human antibody (For example, Riechmann, Nature, 1988, 332, 323-7; Patent Document 1). Rabbit polyclonal antibodies are widely used in biological assays such as ELISA or Western blot. Polyclonal rabbit antibodies are often preferred over polyclonal rodent antibodies because they usually have a much higher affinity. Moreover, rabbits can often elicit good antibody responses to antigens that are poorly immunogenic in mice and / or antigens that do not yield good binders when used in phage display. Because of these well-known advantages of rabbit antibodies, rabbit antibodies are ideal for use in the discovery and development of therapeutic antibodies. The reason this is not generally done is mainly due to technical difficulties in the production of monoclonal rabbit antibodies. Myeloma-like tumors are not known in rabbits and conventional hybridoma technology to generate monoclonal antibodies is not applicable to rabbit antibodies. A pioneering work to provide fusion cell line partners for rabbit antibody expressing cells was performed by Knight et al. (Spieker-Polet et al., PNAS, 1995, 92, 9348-52) and improved fusion partner cell lines were developed. It was described by Pytela et al. In 2005 (see, for example, Patent Document 2). However, this technology is not widely spread because the corresponding know-how is basically controlled by a single research group. Alternative methods for the production of monoclonal antibodies involving cloning of antibodies from selected antibody-expressing cells via RT-PCR have been described in the literature, but the rabbit antibody has never been successfully reported. Absent.

  The rabbit antibody is expected to elicit a strong immune response, similar to the murine antibody, when used in human therapy. Thus, rabbit antibodies need to be humanized before they can be used clinically. However, due to the structural differences between rabbit and mouse antibodies, and between rabbit and human antibodies, respectively, it is not possible to apply the method used to make humanized rodent antibodies to rabbit antibodies. It's not easy. For example, the light chain CDR3 (CDRL3) is often much longer than the previously known human or mouse antibody derived CDRL3.

  Although there are only a few rabbit antibody humanization approaches described in the prior art, they are not a classic fusion approach in which a non-human donor CDR is grafted onto a human acceptor antibody. U.S. Patent No. 6,057,034 describes a so-called "resurfacing" strategy. The goal of the “resurfacing” strategy is to reconstruct solvent accessible residues of non-human frameworks so that they are more human-like. Similar humanization techniques for rabbit antibodies described in US Pat. Both U.S. Patent Nos. 5,099,086 and 5,037,086 disclose methods for humanizing rabbit monoclonal antibodies, which include a comparison of the amino acid sequence of the parent rabbit antibody with the amino acid sequence of a similar human antibody. Subsequently, the amino acid sequence of the parent rabbit antibody is modified such that the framework region of the parent rabbit antibody is more similar in sequence to the equivalent framework region of a similar human antibody. In order to obtain a good binding capacity, it is necessary to make laborious development efforts for each immunobinder individually.

  A potential problem with the above approach is that no human framework is used and the rabbit framework is engineered to make it look more human-like. Such an approach has the risk that the amino acid stretch embedded in the core of the protein may still contain immunogenic T cell epitopes.

  To date, Applicants have not identified a rabbit antibody that has been humanized by applying a fusion approach of the state of the art. This can be explained by the fact that rabbit CDRs can be quite different from human or rodent CDRs. As is known in the art, many rabbit VH chains have an additional counter cysteine compared to their mouse and human counterparts. In addition to the conserved disulfide bridge formed between cys22 and cys92, there is also a cys21-cys79 bridge, and in some rabbit chains, the last residue of CDRH1 and the first residue of CDRH2 There are also S—S bridges between the CDRs formed between them. Moreover, paired cysteine residues are often found in CDR-L3. Furthermore, many rabbit antibody CDRs do not belong to any previously known standard structure. In particular, CDR-L3 is often much longer than its human or mouse counterpart, CDR-L3.

  Therefore, fusion of non-human CDR antibodies into the human framework is a major protein engineering challenge. The transition of the antigen binding loop from a naturally evolved framework to a different artificially selected human framework must be performed such that the natural loop conformation is retained for antigen binding. Often, antigen binding affinity is greatly reduced or lost after loop fusion. In the fusion of antigen binding loops, the use of a carefully selected human framework maximizes the probability of retaining binding affinity in a humanized molecule (Roguzka et al., 1996). Many of the fusion experiments available in the literature provide rough guidance for CDR fusion, but it is impossible to generalize the pattern. A typical problem is losing specificity, stability or productivity after fusing the CDR loops.

US Pat. No. 5,693,761 US Pat. No. 7,429,487 International Publication No. 04/016740 International Publication No. 08/144757 International Publication No. 05/016950

  Thus, there is an urgent need for improved methods for reliably and rapidly humanizing rabbit antibodies for use as therapeutic and diagnostic agents. Furthermore, there is a need for a human acceptor framework to reliably humanize rabbit antibodies, which provides functional antibodies and / or antibody fragments with drug-like biophysical properties.

(Summary of Invention)
Surprisingly, a highly soluble and stable human identified by a quality control (QC) assay (such as disclosed in WO0148017 and Auf der Maur et al. (2001), FEBS Lett, 508, 407-412). It has been found that the antibody framework is particularly suitable for accommodating CDRs from other non-human animal species, such as rabbit CDRs. Thus, in a first aspect, the present invention provides the light and heavy chain variable regions of certain human antibodies (so-called “FW1.4” antibodies). The antibody does not depend on the presence or absence of disulfide bridges in the CDR, and is particularly suitable as a universal acceptor for CDRs from various antibodies of various binding specificities, particularly rabbit antibodies. Furthermore, the present invention provides two variant sequences of the specific human antibody framework, namely rFW1.4 and rFW1.4 (V2). Both are particularly suitable frameworks as versatile acceptor frameworks for fusion of rabbit CDRs. In another aspect, the present invention provides framework residue motifs that provide human frameworks suitable for accommodating CDRs from other non-human animal species, particularly rabbit CDRs.

  Humanized immunobinders generated by fusing rabbit CDRs in these highly compatible variable light and heavy chain frameworks consistently align the spatial orientation of rabbit antibodies (donor CDRs are derived from). And hold it securely. Therefore, it is not necessary to introduce structurally important positions of the donor immunobinder into the acceptor framework. Because of these advantages, humanization of high-throughput rabbit antibodies can be realized with little or no optimization of binding capacity.

Accordingly, in another aspect, the present invention fuses rabbit and other non-human CDRs into the soluble and stable light and / or heavy chain human antibody framework sequences disclosed herein, Provides a method for producing humanized antibodies with superior biophysical properties. In particular, the immunobinder produced by the method of the present invention exhibits excellent functional properties such as solubility and stability.
For example, the present invention provides the following items.
(Item 1)
An immunobinder specific for the desired antigen comprising (i) a rabbit variable light chain CDR, and (ii) a human variable heavy chain framework having at least 50% identity to SEQ ID NO: 4.
(Item 2)
The immunobinder acceptor framework of item 1, wherein the variable heavy chain framework is selected from the group consisting of SEQ ID NO: 4 and SEQ ID NO: 6.
(Item 3)
The immunobinder acceptor framework according to any one of the preceding items, further comprising a variable light chain framework having at least 85% identity to SEQ ID NO: 2.
(Item 4)
4. The immunobinder acceptor framework of item 3, comprising a threonine (T) at position 87 (AHo numbering) of the light chain variable framework.
(Item 5)
The immunobinder acceptor framework according to any one of the preceding items, comprising a variable heavy chain framework and a variable light chain framework, both linked by a linker, in particular the linker of SEQ ID NO: 8.
(Item 6)
A framework having at least 70% identity to SEQ ID NO: 3, SEQ ID NO: 5 or SEQ ID NO: 7, in particular comprising or comprising SEQ ID NO: 3, SEQ ID NO: 5 or SEQ ID NO: 7 The immunobinder acceptor framework according to any one of the preceding items, comprising a framework that is 7.
(Item 7)
Threonine at position 24 (T), valine at position 25 (V), alanine (A) or glycine at position 56 (G), lysine at position 82 (K), threonine at position 84 (T), valine at position 89 ( V) and a human or humanized immunobinder variable heavy chain acceptor framework comprising at least three amino acids from the group consisting of arginine (R) at position 108 (AHo numbering).
(Item 8)
The immunobinder acceptor framework according to any one of the preceding items, wherein the amino acids at positions 12, 103 and / or 144 (AHo numbering) are replaced with hydrophilic amino acids.
(Item 9)
In said heavy chain framework (a) serine (S) at position 12;
(B) serine (S) or threonine (T) at position 103, and / or (c) serine (S) or threonine (T) at position 144
9. The immunobinder acceptor framework according to item 8, having (AHo numbering).
(Item 10)
In the variable light chain framework, glutamic acid at position 1 (E), valine at position 3 (V), leucine at position 4 (L), serine at position 10 (S), arginine at position 47 (R), The immunobinder acceptor framework according to any one of the preceding items having serine (S), phenylalanine (F) at position 91 and / or valine (V) at position 103 (AHo numbering).
(Item 11)
Any of the preceding items further comprising a donor immunobinder, particularly a heavy chain CDR1, CDR2 and CDR3 from a mammalian immunobinder, preferably a rabbit immunobinder, and / or a light chain CDR1, CDR2 and CDR3. The described immunobinder acceptor framework.
(Item 12)
12. The immunobinder acceptor framework according to item 11, further comprising a donor framework residue involved in antigen binding.
(Item 13)
The immunobinder acceptor framework according to any one of the preceding items, further comprising glycine (G) (AHo numbering) at position 141 of the variable heavy chain.
(Item 14)
14. Use of the immunobinder acceptor framework according to any one of items 1-13 for fusion of rabbit CDRs.
(Item 15)
14. An immunobinder comprising the immunobinder acceptor framework according to any one of items 1-13.
(Item 16)
16. The immunobinder according to item 15, which is an scFv antibody.
(Item 17)
One or more binding molecules, particularly therapeutic agents such as cytotoxic agents, cytokines, chemokines, growth factors or other signaling molecules, contrast agents, or second proteins, especially transcription activators or DNA binding domains The immunobinder according to any one of items 15 to 16, further comprising:
(Item 18)
18. Use of an immunobinder according to any one of items 15 to 17 in diagnostic application, therapeutic application, target evaluation or gene therapy.
(Item 19)
A nucleic acid encoding the immunobinder acceptor framework according to any one of items 1 to 13 or the immunobinder according to any one of items 15 to 17, particularly an isolated nucleic acid.
(Item 20)
A vector comprising the nucleic acid according to item 19.
(Item 21)
A host cell comprising the vector according to item 20.
(Item 22)
Use of the nucleic acid according to item 19 or the vector according to item 20 in gene therapy. (Item 23)
The immunobinder acceptor framework according to any one of items 1 to 13, the immunobinder according to any one of items 15 to 17, the nucleic acid according to item 19, or the vector according to item 20. Composition.
(Item 24)
A method of humanizing a rabbit immunobinder, wherein the rabbit immunobinder comprises a heavy chain CDR1, CDR2 and CDR3 sequence, and / or a light chain CDR1, CDR2 and CDR3 sequence, the method comprising:
(I) fusing at least one heavy chain CDR of the group consisting of CDR H1, CDR H2 and CDR H3 sequences within a human heavy chain acceptor framework having at least 50% identity to SEQ ID NO: 1; and (Ii) a human light chain acceptor framework having at least 50% identity to SEQ ID NO: 2, ie, within the human light chain framework, at least one of the group consisting of CDR L1, CDR L2 and CDR L3 sequences Fusing the light chain CDRs.
(Item 25)
(I) the human heavy chain framework comprises the framework amino acid sequence of SEQ ID NO: 1, SEQ ID NO: 4 or SEQ ID NO: 6;
(Ii) the human light chain framework comprises the framework amino acid sequence of SEQ ID NO: 2 or SEQ ID NO: 9;
25. A method according to item 24.
(Item 26)
26. The method according to item 24 or 25, further comprising the step of substituting a framework residue with a framework residue of the rabbit immunobinder involved in antigen binding.
(Item 27)
Preferably by substitution at at least one of the heavy chain amino positions 12, 103 and 144 (AHo numbering), in particular by substitution with hydrophilic amino acids, (a) serine (S), position (b) at position 12 From item 24, further comprising substituting a framework residue of the heavy chain acceptor framework with a substitution selected from the group consisting of 103 threonine (T), and (c) threonine (T) at position 144. 27. The method according to any one of 26.
(Item 28)
A stability enhancing substitution at at least one of positions 1, 3, 4, 10, 47, 57, 91 and 103 of the light chain variable region according to the AHo numbering system, in particular (a) glutamic acid (E) at position 1; (B) valine (V) at position 3, (c) leucine (L) at position 4, (d) serine (S) at position 10, (e) arginine (R) at position 47, (e) at position 57 Substitution of a light chain acceptor framework framework residue with a substitution selected from the group consisting of serine (S), (f) phenylalanine (F) at position 91, and (g) valine (V) at position 103 28. A method according to any one of items 24 to 27, further comprising the step of:
(Item 29)
29. A humanized immunobinder by the method according to any one of items 24-28.
(Item 30)
30. The immunobinder according to item 29, wherein the target antigen is VEGF or TNFα. (Item 31)
a) incubating a rabbit B cell with a target antigen, b) selecting a rabbit B cell that binds to the target antigen, and c) identifying a CDR of an antibody variable domain produced by the B cell. 29. The method of any one of items 24 to 28, further comprising identification of a donor CDR of an antibody variable domain produced by a B cell.
(Item 32)
32. The method of item 31, wherein the B cells are isolated from a rabbit immunized against the target antigen, preferably by DNA vaccination.
(Item 33)
Item 33 or Item 32, wherein the B cells and the target antigen are stained differently and the co-existence of the two stains is positively selected by flow cytometry based single cell sorting. Method.
(Item 34)
34. A method according to any one of items 31 to 33, wherein the target antigen is a soluble protein.
(Item 35)
34. A method according to any one of items 31 to 33, wherein the target antigen is expressed on the surface of a cell, in particular the target antigen is a transmembrane protein, in particular a multi-transmembrane protein.
(Item 36)
36. The method of item 35, wherein the target antigen is indirectly stained by staining cells expressing the target antigen with an intracellular fluorescent dye.
(Item 37)
37. A method according to any one of items 31 to 36, wherein the B cells are stained with one or more labeled antibodies specific for a B cell surface marker.
(Item 38)
38. A method according to any one of items 31 to 37, wherein the B cells are stained with a labeled anti-IgG antibody.
(Item 39)
40. The method of item 38, wherein the B cells are stained with a labeled anti-IgG antibody and a differently labeled anti-IgM antibody.
(Item 40)
40. The method of item 39, wherein B cells stained with the labeled anti-IgM antibody are selected negatively.
(Item 41)
From item 31 further comprising culturing said B cell selected in step b) of item 31 such that said generated antibody is secreted into the medium, particularly in the presence of a helper cell line 41. The method according to any one of 40.
(Item 42)
42. The method of any one of items 41, further comprising the step of testing the secreted antibody for specific binding to the target antigen.
(Item 43)
Item wherein the CDRs of the antibody are amplified, in particular by RT-PCR, fused into an acceptor framework to produce an immunobinder and subsequently tested for specific binding to the target antigen. 43. The method according to 42.
(Item 44)
For fusion of rabbit CDRs of an immunobinder acceptor framework comprising a sequence having at least 90% identity to SEQ ID NO: 1 and / or a sequence having at least 85% identity to SEQ ID NO: 2. use.
(Item 45)
45. Use according to item 44, wherein the framework comprises SEQ ID NO: 1 and SEQ ID NO: 2, both connected by a linker, in particular by a linker having SEQ ID NO: 8.

FIG. 1 shows the definition of CDR H1 used here to fuse an antigen binding site from a rabbit monoclonal antibody into a highly soluble and stable human antibody framework. FIG. 2 schematically shows B cell 1 labeled with fluorescent antibody 2 that interacts with target-expressing cell 3 stained with intracellular dye 4. Selected targets: 5; BCR: 6. FIG. 3 shows the FACS selection process of rabbit B cells binding to ESBA903 soluble target. FIG. 3A: Lymphocytes are gated by forward and side scatter. FIG. 3B: Among them, IgG + IgM− cells (probably memory B cells) are selected (red gate). FIG. 3C: Cells double stained with ESBA903-PE and ESBA903-PerCP (green gate) are predicted to encode high affinity IgG for ESBA903. Cells that showed the brightest fluorescence (pink gate) were sorted into 96-well plates. FIG. 4 shows that beads coated with anti-TNFα antibody (PE label) bind to TNFα-transfected CHO cells (upper panel), control beads coated with anti-CD19 antibody (APC label) are TNFα-transfected CHO cells. No binding to (middle panel), beads coated with anti-TNFα antibody (PE labeled) do not bind to wild type (wt) CHO cells (lower panel). The dot plot on the left shows forward and side scatter, which show the event size and granularity, respectively. A single bead (approximately 3 μm) population is gated to P2. CHO cells (about 30 μm) finally bound to the beads are gated to P1. The middle dot plot shows P1 events (CHO cells) for PE or APC staining. Thus, if cells interact with anti-TNFα beads, they are shown at the P3 gate, and if they interact with anti-CD19 beads, they appear at the P4 gate. On the right side, the statistical details for each sample are shown. FIG. 5 shows that beads coated with anti-TNFα-PE and beads coated with anti-CD19APC were mixed with TNFα-transfected CHO cells and CHO cells were gated (P1), among them, coated with anti-TNFαPE Cells that bind to either the bead or the anti-CD19-APC coated bead are shown at gates P3 and P4, respectively, indicating that unbound beads are visible at gate P2. FIG. 6 is an analysis of rabbit antibody sequences extracted from the Kabat database, which confirms that the CDR3 of the variable heavy chain is usually 3 amino acids longer than its mouse counterpart.

(Detailed description of the invention)
(Definition)
In order that the present invention may be more readily understood, certain terms are defined as follows. Additional definitions are set forth throughout this detailed description.

The term “antibody” refers to intact antibodies and any antigen binding fragment. The terms “antigen-binding polypeptide” and “immunobinder” are used simultaneously herein. “Antibody” refers to a protein or antigen-binding portion thereof comprising at least two heavy (H) chains and two light (L) chains interconnected by disulfide bonds, optionally glycosylated. Each heavy chain includes a heavy chain variable region (abbreviated herein as V _H ) and a heavy chain constant region. The heavy chain constant region includes three domains, CH1, CH2 and CH3. Each light chain includes a light chain variable region (abbreviated herein as _VL ) and a light chain constant region. The light chain constant region includes one domain, CL. The V _H and V _L regions can be further subdivided into hypervariable regions called complementarity determining regions (CDRs) and more conserved regions called framework regions (FR) interspersed with hypervariable regions. Each V _H and V _L is composed of three CDRs and four FRs, arranged from the amino terminus to the carboxy terminus in the following order: FR1, CDR1, FR2, CDR2, FR3, CDR3, FR4. The variable region of the heavy and light chains contains a binding domain that interacts with an antigen. The constant region of the antibody may mediate immunoglobulin binding to host tissues or factors, including various cells of the immune system (eg, effector cells) and the first component of the classical complement system (Clq).

The term “antigen-binding portion” (or simply “antibody portion”) of an antibody refers to one or more fragments of an antibody that retain the ability to specifically bind to an antigen (eg, TNF). It has been shown that the antigen-binding function of an antibody can be performed by fragments of a full-length antibody. Examples of binding fragments encompassed by the term “antigen-binding portion” of an antibody include: (i) a Fab fragment that is a monovalent fragment consisting of V _L , V _H , CL and CH1 domains; (ii) a disulfide in the hinge region F (ab ′) ₂ fragment, which is a bivalent fragment comprising two Fab fragments linked by cross-linking, (iii) Fd fragment consisting of V _H and CH1 domains, (iv) single arm _VL and V _H of the antibody Fv fragment consisting of a domain, (v) a single domain or dAb fragment consisting of one _VH domain (Ward et al. (1989), Nature, 341: 544-546), and (vi) isolated complementarity Optionally by a determining region (CDR), or (vii) a synthetic linker Examples include combinations of two or more isolated CDRs that may be linked. Furthermore, the two domains of the Fv fragment, namely _VL and _VH , are encoded by separate genes, but they use recombinant methods with synthetic linkers that allow them to be made as a single protein chain. In this single protein chain, the V _L and V _H regions are paired and known as a monovalent molecule (single chain Fv (scFv); see, eg, Bird et al. (1988) Science, 240, 423-426; and Huston et al. (1988) Proc. Natl. Acad. Sci. USA, 85, 5879-5883). Such single chain antibodies are also intended to be encompassed within the term “antigen-binding portion” of an antibody. These antibody fragments are obtained using conventional techniques known to those skilled in the art and are screened for utility in the same manner as intact antibodies. The antigen binding portion may be produced by recombinant DNA techniques or may be produced by enzymatic or chemical cleavage of intact immunoglobulins. The antibody can be of a different isotype, eg, an IgG (eg, IgG1, IgG2, IgG3 or IgG4 subtype), IgA1, IgA2, IgD, IgE or IgM antibody.

The term “immunobinder” refers to a molecule that contains all or part of an antigen binding site of an antibody, eg, all or part of a heavy chain variable domain and / or light chain variable domain, and thus an immunobinder is Specific recognition of target antigen. Non-limiting examples of immunobinders include, but are not limited to, full-length immunoglobulin molecules and scFv, (i) Fab fragments: _VL domain, V _H domain, C _L domain and C _H 1 A monovalent fragment consisting of a domain; (ii) a F (ab ′) ₂ fragment, ie a divalent fragment comprising two Fab fragments joined by a disulfide bridge in the hinge region; (iii) essentially Fab ′ fragment, which is a Fab with part of the hinge region (see FUNDAMENTAL IMMUNOLOGY (Paul, 3rd edition, 1993)); (iv) Fd fragment consisting of V _H domain and C _H 1 domain; v) the single arm _VL domain of the antibody and Fv fragment consisting of _VH domain; (vi) Dab fragment consisting of _VH domain or _VL domain (Ward et al., (1989) Nature 341: 544-546), camel antibody (Hamers-Casterman et al., Nature) 363: 446-448 (1993) and Dumoulin et al., Protein Science 11: 500-515 (2002), or shark antibodies (eg, Shark Ig-NAR Nanobodies®). And (vii) antibody fragments, including a heavy chain variable region containing a single variable domain and two constant domains, Nanobodies.

The terms “single chain antibody”, “single chain Fv”, or “scFv” refer to an antibody heavy chain variable domain (or region; V _H ) and an antibody light chain variable domain (or region; V _L ) connected by a linker. ). Such scFv molecules with the general structure: _NH 2 -V _L - linker _-V H -COOH or _NH 2 -V _H - may have a linker _-V L -COOH. A suitable linker in the state of the art consists of a repetitive GGGGS amino acid sequence or a variant thereof. In a preferred embodiment of the present invention, a (GGGGGS) ₄ linker of the amino acid sequence set forth in SEQ ID NO: 8 is used, although 1 to 3 repeat variants are possible (Holliger et al. (1993) Proc. Natl. Acad. Sci. USA, 90, 6444-6448). Other linkers that can be used in the present invention are described in Alfthan et al. (1995), Protein Eng. 8, 725-731, Choi et al. (2001), Eur. J. et al. Immunol. 31, 94-106, Hu et al. (1996), Cancer Res. 56, 3055-3061, Kipriyanov et al. (1999), J. Am. Mol. Biol. 293, 41-56 and Roovers et al. (2001), Cancer Immunol. It is described by.

  As used herein, the term “functional property” is used by those of ordinary skill in the art, for example, to improve the manufacturing properties or therapeutic effectiveness of a polypeptide (eg, compared to a conventional polypeptide). A property of a polypeptide (eg, immunobinder) for which improvement is desirable and / or advantageous. In one embodiment, the functional property is stability (eg, thermal stability). In another embodiment, the functional property is solubility (eg, under cellular conditions). In yet another embodiment, the functional property is agglomeration behavior. In yet another embodiment, the functional property is protein expression (eg, in prokaryotic cells). In yet another embodiment, the functional property is refolding behavior after solubilization of inclusion bodies in the manufacturing process. In certain embodiments, the functional property is not an improvement in antigen binding affinity. In another preferred embodiment, the improvement in one or more functional properties does not substantially affect the binding affinity of the immunobinder.

  The term “CDR” refers to one of the six hypervariable regions within the variable domain of an antibody that primarily contribute to antigen binding. One of the most commonly used definitions for the six CDRs is Kabat E.I. A. (1991) Sequences of proteins of immunological interest. , NIH Publication, pages 91- 3242. As used herein, CDR Kabat definition includes CDR1, CDR2, and CDR3 (CDR L1, CDR L2, CDR L3, or L1, L2, L3) of light chain variable domains and heavy chain variable domains. Only applies to CDR2 and CDR3 (CDR H2, CDR H3, or H2, H3). However, as used herein, the heavy chain variable domain CDR1 (CDR H1 or H1) is defined by the position of the residue beginning at position 26 and ending before position 36 (Kabat numbering). This is basically a fusion of CDR H1 defined differently by Kabat and Chotia (see also FIG. 1 for illustration).

  As used herein, the term “antibody framework” means the portion of the variable domain of either VL or VH that serves as the backbone for the antigen-binding loop (CDR) of the variable domain. In essence, this is a variable domain without CDRs.

  The term “epitope” or “antigenic determinant” refers to a site on an antigen to which an immunoglobulin or antibody specifically binds (eg, a specific site on a TNF molecule). Epitopes usually contain at least 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14 or 15 consecutive or non-contiguous amino acids in a unique spatial conformation . For example, “Epitope Mapping Protocols in Methods in Molecular Biology”, Vol. E. See edited by Morris (1996).

The terms “specific binding”, “selective binding”, “selectively bind” and “specifically bind” refer to antibody binding to an epitope of a given antigen. Usually, the antibody binds with an affinity (K _D ) of less than about 10 ⁻⁷ M, such as less than about 10 ⁻⁸ M, less than about 10 ⁻⁹ M, or less than about 10 ⁻¹⁰ M or even lower values.

The term “K _D ” or “Kd” refers to the dissociation equilibrium constant of a particular antibody-antigen interaction. Typically, antibodies of the invention are less than about 10 ⁻⁸ M, less than about 10 ⁻⁹ M, or less than about 10 ⁻¹⁰ M or even lower, as determined using, for example, the surface plasmon resonance (SPR) technique of the BIACORE instrument It binds to TNF with a dissociation equilibrium constant (K _D ) of less than about 10 ⁻⁷ M, such as value.

  The term “nucleic acid molecule” as used herein refers to DNA and RNA molecules. The nucleic acid molecule may be single-stranded or double-stranded, but preferably is double-stranded DNA. A nucleic acid is “operably linked” when it is placed into a functional relationship with another nucleic acid sequence. For example, a promoter or enhancer is operably linked to a sequence if it affects the transcription of the coding sequence.

  The term “vector” refers to a nucleic acid molecule capable of transporting another nucleic acid to which it has been linked. In one embodiment, the vector is a “plasmid”, which refers to a circular double stranded DNA loop into which additional DNA segments can be ligated. In another embodiment, the vector is a viral vector, in which case additional DNA segments can be ligated into the viral genome. The vectors disclosed herein can be capable of autonomous replication in the host cell into which they are introduced (eg, bacterial vectors having a bacterial origin of replication and episomal mammalian vectors). Alternatively, it can be one that is integrated into the genome of the host cell upon introduction into the host cell and thereby can be replicated along with the host genome (eg, a non-episomal mammalian vector).

The term “host cell” refers to a cell into which an expression vector has been introduced. Host cells include bacterial cells, microbial cells, plant cells or animal cells, preferably Escherichia coli, Bacillus subtilis; Saccharomyces
cerevisiae, Pichia pastoris, CHO (Chinese hamster ovary cell line) or NS0 cells.

  The term “lagomorph” refers to members of the taxonomic Lagomorpha family and includes the Leporidae family (eg, hares and rabbits) and the Ochotonidae family (Nara rabbit). In the most preferred embodiment, the rabbit is a rabbit. The term “rabbit” as used herein refers to an animal belonging to the family Lepididae.

  As used herein, “identity” means a sequence that matches between two polypeptides, molecules, or two nucleic acids. If a position in both of the two sequences compared is occupied by the same base or the same amino acid monomer subunit (eg, if a position in each of the two polypeptides is occupied by lysine), respectively Are identical at that position. "Percent identity" between two sequences is shared by those sequences, taking into account the number of gaps that need to be introduced for optimal alignment of the two sequences and the length of each gap It is a function of the number of identical positions. In general, a comparison is made when two sequences are aligned to give the greatest identity. Such an alignment may include, for example, either a Blossum 62 matrix or a PAM 250 matrix, and gap weights 16, 14, 12, 10, 8, 6, or 4 and length weights 1, 2, 3, 4, 5, or 6. And can be provided using the algorithmic method of Needleman and Wunsch (J. Mol. Biol., (48), 444-453 (1970)) incorporated into the GAP program in the GCG software package. .

  Unless defined otherwise, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention belongs. Although methods and materials similar or equivalent to those described herein can be used in the practice or testing of the present invention, suitable methods and materials are those described below. In case of conflict, the present specification, including definitions, will control. In addition, the materials, methods, and examples are illustrative only and not intended to be limiting.

  Various aspects of the invention are described in further detail in the following subsections. It is understood that various embodiments, selections, and ranges can be arbitrarily combined. Further, in certain embodiments, selected definitions, embodiments, or ranges may not apply.

  Unless otherwise noted, amino acid positions are indicated according to the AHo numbering scheme. The AHo numbering system is described by Honegger, A .; And Pluckthun, A .; (2001), J.M. Mol. Biol. 309, 657-670). Instead, Kabat et al. (Kabat, E. A. et al. (1991), “Sequences of Proteins of Immunological Interest.”, 5th edition, US Department of Health and Human Services number I-3, 24, published by N.S. The Kabat numbering system described in more detail can be used. A conversion table of two different numbering systems used to identify amino acid residue positions within antibody heavy and light chain variable regions is shown in A.C. Honegger, J. et al. Mol. Biol. 309 (2001), pages 657-670.

  In a first aspect, the present invention provides a universal acceptor framework for fusing CDRs from other animal species such as rabbits. It has been previously described that antibodies or antibody derivatives comprising human frameworks identified in so-called “quality control” screening (WO0148017) are generally characterized by high stability and / or solubility. The human single chain framework FW1.4 (a combination of SEQ ID NO: 1 (designated a43 in WO 03/097697) and SEQ ID NO: 2 (designated K127 in WO 03/097697)) is used in quality control assays. Although it was clearly poor, it was surprising to see that this framework has a high degree of intrinsic thermodynamic stability and good reproducibility when combined with various different CDRs. It was issued. The stability of this molecule can largely be attributed to its framework region. It has further been shown that FW1.4 is inherently highly compatible with the antigen binding site of the rabbit antibody. Thus, FW1.4 is a suitable scaffold for constructing stable humanized scFv antibody fragments obtained from fusion of rabbit loops. Thus, in one aspect, the invention comprises a VH sequence having at least 90% identity to SEQ ID NO: 1 and / or a VL sequence having at least 85% identity to SEQ ID NO: 2, more preferably a rabbit CDR FW1.4 sequence for fusion of (SEQ ID NO: 3), or SEQ ID NO: 3 at least 60%, more preferably at least 65%, 70%, 75%, 80%, 85%, 90%, 95% Provided is an immunobinder acceptor framework comprising sequences with identity.

  Further, it has been found that FW1.4 can be optimized by replacing several residue positions in the heavy chain of FW1.4 and / or by replacing one position in the light chain of FW1.4. It was done. Thereby, it was surprisingly found that the loop conformation of the extensive rabbit CDRs in VH can be fully maintained, largely independent of the donor framework sequence. The residue in the heavy chain of FW1.4 and one position in the light chain are conserved in rabbit antibodies. A consensus residue at that position in the heavy chain and one position in the light chain was deduced from the rabbit repertoire and introduced into the sequence of the human acceptor framework.

  As a result, the modified framework 1.4 (hereinafter referred to as rFW1.4) is compatible with virtually any rabbit CDR. Furthermore, rFW1.4 containing various rabbit CDRs is well expressed, well produced, and almost completely retains the affinity of the original donor rabbit antibody, as opposed to the rabbit wild type single chain. To do.

  Accordingly, the present invention provides a variable heavy chain framework of SEQ ID NO: 1 further comprising one or more amino acid residues that generally support the conformation of CDRs from a rabbit immunobinder. In particular, the residue is present at one or more amino acid positions selected from the group consisting of 24H, 25H, 56H, 82H, 84H, 89H and 108H (AHo numbering). These positions have been found to affect CDR conformation, and thus mutations to accommodate donor CDRs are contemplated. The one or more residues include threonine at position 24 (T), valine at position 25 (V), glycine or alanine at position 56 (G or A), lysine at position 82 (K), threonine at position 84 (T), preferably selected from the group consisting of valine (V) at position 89 and arginine (R) at position 108 (AHo numbering). It is preferred that there are at least 3 residues, more preferably 4, 5, 6 residues, most preferably all 7 residues. Surprisingly, it has been found that the presence of the above mentioned residues improves the stability of the immunobinder.

  In a preferred embodiment, the invention provides a threonine (T) at position 24, a valine (V) at position 25, an alanine (A) or glycine (G) at position 56, a threonine (T) at position 84, a lysine at position 82. (K), valine (V) at position 89, and arginine (R) at position 108 (AHo numbering), at least 1 residue, more preferably at least 3 residues, more preferably 4, 5, 6 At least 50%, more preferably at least 55%, 60%, 65%, 70%, 75%, 80%, 85% in SEQ ID NO: 4, provided that there are residues, most preferably 7 residues, An immunobinder acceptor framework comprising VH having 90%, 95%, and even more preferably 100% identity is provided. In a preferred embodiment, the immunobinder acceptor framework is an immunobinder acceptor framework for rabbit CDRs.

  In a preferred embodiment, the variable heavy chain framework is SEQ ID NO: 4 or SEQ ID NO: 6, or comprises SEQ ID NO: 4 or SEQ ID NO: 6. Both the variable heavy chain frameworks can be combined with any suitable light chain framework, for example.

Therefore, the present invention
(I) a variable heavy chain framework having at least 70%, preferably at least 75%, 80%, 85%, 90%, more preferably at least 95% identity to SEQ ID NO: 4, and / or (ii) the sequence An immunobinder acceptor framework comprising a variable light chain framework having at least 70%, preferably at least 75%, 80%, 85%, 90%, more preferably at least 95% identity to No. 2 is provided.

  In a highly preferred embodiment, the variable heavy chain framework is threonine at position 24 (T), glycine at position 56 (G), threonine at position 84 (T), valine at position 89 (V), and arginine at position 108 (R ) (AHo numbering).

  In a preferred embodiment, the variable light chain comprises threonine (T) at position 87 (AHo numbering).

In a preferred embodiment, the immunobinder acceptor framework is
(I) a variable heavy chain framework selected from the group consisting of SEQ ID NO: 1, SEQ ID NO: 4 and SEQ ID NO: 6, and / or (ii) a variable light chain framework of SEQ ID NO: 2 or SEQ ID NO: 9.

In a preferred embodiment, the variable heavy chain framework is linked to the variable light chain framework by a linker. The linker can be any suitable linker, for example a linker comprising the sequence GGGGS of 1 to 4 repeats, preferably (GGGGS) ₄ peptide (SEQ ID NO: 8), or Alfthan et al. (1995), Protein Eng. 8, 725-731.

  In another preferred embodiment, the immunobinder acceptor framework is a sequence having at least 70%, 75%, 80%, 85%, 90%, more preferably at least 95% identity to SEQ ID NO: 5; On the other hand, the sequence is preferably not SEQ ID NO: 3. More preferably, the immunobinder acceptor framework comprises or is SEQ ID NO: 5.

  In another preferred embodiment, the immunobinder acceptor framework is a sequence having at least 70%, 75%, 80%, 85%, 90%, more preferably at least 95% identity to SEQ ID NO: 7; On the other hand, the sequence is preferably not SEQ ID NO: 3. More preferably, the immunobinder acceptor framework comprises or is SEQ ID NO: 7.

  Furthermore, surprisingly, the presence of the above-mentioned amino acid motifs has been found to make the framework, preferably the human framework, particularly suitable for accommodating CDRs from other non-human animal species, particularly rabbit CDRs. It was issued. The motif does not have a negative impact on the stability of the immunobinder. The CDRs appear in a conformation similar to their natural spatial orientation in the rabbit immunobinder, and thus no structurally important positions need to be fused to the acceptor framework. Thus, human or humanized immunobinder acceptor frameworks include threonine (T) at position 24, valine (V) at position 25, alanine (A) or glycine (G) at position 56, lysine (K) at position 82. , Threonine at position 84 (T), valine at position 89 (V) and arginine at position 108 (R) (AHo numbering), preferably at least 3 amino acids, preferably 4, 5, 6 amino acids, more preferably 7 Contains amino acids.

  The immunobinder acceptor framework described herein may include substitutions that promote solubility at the heavy chain framework, preferably at positions 12, 103 and 144 (AHo numbering). Hydrophobic amino acids are preferably substituted with more hydrophilic amino acids. Examples of hydrophilic amino acids include arginine (R), asparagine (N), aspartic acid (D), glutamine (Q), glycine (G), histidine (H), lysine (K), serine (S) and threonine. (T). More preferably, the heavy chain framework is (a) serine (S) at position 12, (b) serine (S) or threonine (T) at position 103, and / or (c) serine (S) at position 144. Alternatively, it contains threonine (T).

  In addition, there may be a stability enhancing amino acid at one or more of positions 1, 3, 4, 10, 47, 57, 91 and 103 (AHo numbering) of the variable light chain framework. The variable light chain framework consists of glutamic acid at position 1 (E), valine at position 3 (V), leucine at position 4 (L), serine at position 10 (S); arginine at position 47 (R), position 57 More preferably, it comprises serine (S), phenylalanine (F) at position 91 and / or valine (V) at position 103.

  In another preferred embodiment, VH contains glycine (G) at position 141 because glutamine (Q) is prone to deamination. This substitution can improve long-term storage of the protein.

For example, the acceptor frameworks disclosed herein can be used to generate human or humanized antibodies that retain the binding properties of non-human antibodies from which non-human CDRs are derived. Thus, in a preferred embodiment, the invention provides a donor immunobinder, preferably a mammalian immunobinder, more preferably a rabbit immunobinder, most preferably a heavy chain CDR1, CDR2 and CDR3 from a rabbit, and / or a light chain CDR1, The immunobinder acceptor framework disclosed herein further comprising CDR2 and CDR3. Thus, in one embodiment, the present invention provides
(I) a rabbit variable light chain CDR;
(Ii) at least 50%, more preferably at least 55%, 60%, 65%, 70%, 75%, 80%, 85%, 90%, 95%, most preferably 100% identity to SEQ ID NO: 4 An immunobinder specific for the desired antigen is provided comprising a human variable heavy chain framework having

  In a preferred embodiment, the human variable heavy chain framework sequence includes a threonine at position 24 (T), a valine at position 25 (V), an alanine at position 56 (A) or a glycine (G), a threonine at position 84. There is a condition that at least one amino acid of the group consisting of (T), lysine (K) at position 82, valine (V) at position 89 and arginine (R) at position 108 (AHo numbering) is present.

  Preferably, the rabbit is a rabbit. More preferably, the immunobinder comprises a heavy chain CDR1, CDR2 and CDR3 derived from a donor immunobinder and a light chain CDR1, CDR2 and CDR3.

  As is known in the art, many rabbit VH chains have an additional counter cysteine compared to their mouse and human counterparts. In addition to the conserved disulfide bridge formed between cys22 and cys92, there is also a cys21-cys79 bridge, and in some rabbit chains, the last residue of CDRH1 and the first residue of CDRH2 There are also S—S bridges between the CDRs formed between them. Furthermore, cysteine residues are often found in CDR-L3. Furthermore, many rabbit antibody CDRs do not belong to any previously known standard structure. In particular, CDR-L3 is often much longer than its human or mouse counterpart, CDR-L3.

As described above, fusion of a non-human CDR to the framework disclosed herein yields a molecule in which the CDR appears in an appropriate conformation. If necessary, the affinity of the immunobinder can be improved by fusing the antigen interaction framework residues of the non-human donor immunobinder. These positions are, for example,
(I) identifying each germline precursor sequence, or instead using a consensus sequence in the case of highly homologous framework sequences;
(Ii) generating a sequence alignment of the germline precursor sequence or consensus sequence of step (i) with the donor variable domain sequence, and (iii) identifying the different residues.

  Different residues on the surface of the molecule were often mutated during the in vivo affinity generation process, presumably to generate affinity for the antigen.

  In another aspect, the present invention provides an immunobinder comprising an immunobinder acceptor framework as described herein. For example, the immunobinder can be a scFv antibody, full-length immunoglobulin, Fab fragment, Dab or nanobody.

  In preferred embodiments, the immunobinder is one or more molecules, eg, therapeutic agents such as cytotoxic agents, cytokines, chemokines, growth factors or other signaling molecules, contrast agents, or transcriptional activators or DNA binding. It is bound to a second protein such as a domain.

  The immunobinders disclosed herein can be used, for example, in diagnostic applications, therapeutic applications, target evaluation or gene therapy.

  The present invention further provides an isolated nucleic acid encoding the immunobinder acceptor framework disclosed herein or the immunobinder (s) disclosed herein.

  In another embodiment, a vector comprising a nucleic acid disclosed herein is provided.

  The nucleic acids or vectors disclosed herein can be used, for example, in gene therapy.

  The invention further encompasses host cells comprising the vectors and / or nucleic acids disclosed herein.

  Further provided is a composition comprising an immunobinder acceptor framework disclosed herein, an immunobinder disclosed herein, an isolated nucleic acid disclosed herein, or a vector disclosed herein. .

  The sequences disclosed herein are as follows (X residue is the CDR insertion site):

In another aspect, the invention provides a method of humanizing a non-human antibody by fusing the CDRs of the non-human donor antibody to a stable and soluble antibody framework. In a particularly preferred embodiment, the CDR stem and framework from the rabbit antibody are as described above.

  A general method for fusing CDRs in a human acceptor framework is disclosed by Winter in US Pat. No. 5,225,539 and in Queen, et al. In WO 9007861 A1, which are hereby incorporated by reference in their entirety. The general strategy for fusing CDRs from rabbit monoclonal antibodies to selected frameworks is related to that of Winter et al. And Queen et al., But differs in certain major respects. In particular, the methods of the present invention are typical of the typical Winter and Queen known in the art in that the human antibody frameworks disclosed herein are particularly suitable as universal acceptors for human or non-human donor antibodies. Different from methodology. Thus, unlike the general methods of Winter and Queen, the framework sequences used in the humanization methods of the present invention are not necessarily the best sequence similarity to the sequence of the non-human (eg, rabbit) antibody from which the donor CDRs are derived. It is not a framework array showing sex. In addition, framework residue fusion from the donor sequence is not required to support the CDR conformation. At most, antigen binding amino acids located in the framework or other mutations that occurred during somatic hypermutation can be introduced.

  Specific details of the fusion method for producing humanized rabbit-derived antibodies with high solubility and stability are described below.

Accordingly, the present invention provides a method for humanizing rabbit CDR donor immunobinders comprising heavy chain CDR1, CDR2 and CDR3 sequences and / or light chain CDR1, CDR2 and CDR3 sequences. This method
(I) at least 50%, preferably at least 55%, 60%, 65%, 70%, 75%, 80%, 85%, 90%, more preferably at least 95% identical to SEQ ID NO: 1 on the heavy chain Fusing at least one, preferably two, more preferably three CDRs of the group consisting of CDR1, CDR2 and CDR3 sequences in a human heavy chain acceptor framework having sex; and / or (ii) On the light chain has at least 50%, preferably at least 55%, 60%, 65%, 70%, 75%, 80%, 85%, 90%, more preferably at least 95% identity to SEQ ID NO: 2. In the human light chain acceptor framework, at least one, preferably two, more preferred of the group consisting of CDR1, CDR2, and CDR3 sequences Or a step of fusing three CDRs.

  In a preferred embodiment, the variable chain acceptor framework comprises (i) a human heavy chain framework comprising a framework amino acid sequence selected from the group consisting of SEQ ID NO: 1, SEQ ID NO: 4, and SEQ ID NO: 6; And a human light chain framework comprising the framework amino acid sequence of SEQ ID NO: 2 or SEQ ID NO: 9.

  In a highly preferred embodiment, the method comprises an immunobinder having at least 75%, 80%, 85%, 90%, more preferably at least 95% identity to SEQ ID NO: 3, SEQ ID NO: 5 or SEQ ID NO: 7, (I) fusing heavy chain CDR1, CDR2 and CDR3 sequences in the heavy chain; and (ii) fusing light chain CDR1, CDR2 and CDR3 sequences in the light chain. More preferably, the immunobinder is SEQ ID NO: 3, SEQ ID NO: 5 or SEQ ID NO: 7, or comprises SEQ ID NO: 3, SEQ ID NO: 5 or SEQ ID NO: 7.

  In another embodiment, in order to improve antigen binding, the method may further comprise replacing acceptor framework residues with donor residues involved in antigen binding.

  In an exemplary embodiment of the methods of the invention, the amino acid sequence of a CDR donor antibody is first identified and the sequence is aligned using conventional sequence alignment tools (eg, Needleman-Wunsch algorithm and Blossum matrix). Introduction of gaps and naming of residue positions can be performed using conventional antibody numbering systems. For example, an AHo numbering system can be used for immunoglobulin variable domains. The Kabat numbering scheme is also applicable because it is the most widely adopted standard method for numbering residues in antibodies. Kabat numbering can be assigned using, for example, the SUBIM program. This program analyzes the variable region of antibody sequences and numbers the sequences according to the system established by Kabat et al. (Deret et al., 1995). Framework and CDR regions are usually defined according to the Kabat definition, which is based on sequence variability and is most commonly used. However, for CDR-H1, such designation is the Kabat definition, mean contact data generated by analysis of contacts between antibodies and antigens of a subset of the 3D complex structure (MacCallum et al., 1996), And a combination of Chotia definitions based on the location of the structural loop region (see also FIG. 1). A conversion table of two different numbering systems used to identify the amino acid residue positions of the variable regions of antibody heavy and light chains is shown in A.C. Honegger, J. et al. Mol. Biol. 309 (2001), pages 657-670. The Kabat numbering system is described in Kabat et al. (Kabat, EA et al. (1991), Sequences of Proteins of Immunological Interest, 5th edition, US Department of Health and Humans 24). Are described in more detail. The AHo numbering system is described by Honegger, A .; And Pluckthun, A .; (2001), J.M. Mol. Biol. 309, pp. 657-670.

  The variable domains of rabbit monoclonal antibodies can be classified, for example, into the corresponding human subgroups using classification methods based on, for example, the EXCEL implementation of the sequence analysis algorithm and analysis of the human antibody repertoire (Knappik et al. 2000, J Mol Biol., February 11, 296 (1), 57-86).

  CDR conformations can be assigned to donor antigen binding regions and subsequently the residue positions necessary to maintain the various canonical structures can also be identified. The standard structure of five CDRs out of the six antibody hypervariable regions (L1, L2, L3, H1 and H2) of the rabbit antibody is determined using the definition of Chothia (1989).

  In a preferred embodiment, CDRs are generated, identified and isolated according to the following method: B cells (preferably rabbit B cells) are either (i) target antigen (preferably purified) or (ii) on the surface. Incubate with cells expressing the target antigen.

  In the case of (ii), the cell expressing the target antigen is, for example, a mammalian cell that is naturally expressed for the selected target or is transformed to express the target protein on the surface, preferably CHO or It may be HEK293 cells, yeast cells, preferably yeast spheroplasts, or bacterial cells. Upon expression, the target antigen can be incorporated into the cell membrane or bound to the cell membrane and expressed on the cell surface. The cells may be cultivated, for example, as an isolate in cell culture or isolated from their natural environment, such as a tissue, organ or organism.

  When the target antigen is expressed on the cell surface, ie in (ii), a transmembrane protein is particularly preferred, a multi-transmembrane protein such as a GPCR (G protein coupled receptor) or ion channel, or recombinant expression and Even more preferred are any other proteins that are difficult to maintain their native conformation during purification. Traditional immunization with recombinant proteins is not recommended in these cases due to loss of the native conformation of the integral membrane protein / complex during the purification process or due to insufficient amount of pure protein Or impossible. In a preferred embodiment of the invention, a mammal, more preferably a rabbit, is immunized with DNA instead of the recombinant protein, for example by the DNA vaccination protocol disclosed in WO / 2004/087216. DNA vaccination induces a rapid immune response against the natural antigen. Since no recombinant protein is required, this technique is extremely cost-effective on the one hand and, more importantly, the method is based on integral membrane complexes and / or multiple transmembranes. Allows natural expression of membrane proteins. B cells can be isolated from the immunized mammal, preferably the rabbit, or alternatively can be naive B cells.

In subsequent steps of the method, B cells, preferably memory B cells, are isolated from lymphatic vessels (such as spleen or lymph nodes) of the immunized animal, preferably the immunized rabbit. These B cells are incubated in cells expressing the antigen on the surface or in a mixture with soluble fluorescently labeled antigen. Target-specific antibodies are expressed on their surface, and as a result, B cells that bind to the target antigen or target antigen expressed on the cell surface are isolated. In a highly preferred embodiment, B cells and / or target cells are stained to allow isolation by flow cytometry-based B cell / target cell complex or B cell / antigen complex sorting. Flow cytometry typically measures the fluorescence emitted from a single cell as it crosses the laser beam. However, some researchers already have cell-cell interactions such as cadherin (Panchan et al., 2006, J. Cell Science, 119, 66-74; Leong et Hibma, 2005, J. Biol.
Immunol. Methods, 302, 116-124) or cytometers to investigate adhesion mediated by integrins (Gawaz et al., 1991, J. Clin. Invest, 88, 1288-1134). Yes. Therefore, in the case of (ii), cells expressing the selected target are stained with an intracellular fluorescent dye (for example, calcein). B cells are stained with a fluorescent antibody that binds to a cell surface specific marker. Thus, it is possible to select two-color “events” present in two cells that are attached to each other via B cell receptor-target interactions (see FIG. 2).

  Since IgG usually has a higher affinity than IgM, it is preferred to select positive B cells that express IgG on their surface rather than IgM (which is unique to memory B cells). For this purpose, it is preferred to use multicolor staining in which an antibody specific for IgG and an antibody specific for IgM are differentially labeled, for example with APC and FITC, respectively.

  In one particular embodiment, the read-out for B cell sorting can also be selected for the ability of the interaction to functionally block or activate receptor signaling. For example, B cells can be incubated with cells that functionally express a GPCR (G protein coupled receptor). An agonist that signals via GPCR can be added to the mixture to induce Ca2 + efflux from the GPCR-mediated endoplasmic reticulum. When antibodies presented to B cells functionally block agonist signaling, the result is that Ca 2+ efflux is also blocked by this intercellular interaction. Ca2 + efflux can be measured quantitatively by flow cytometry. Thus, only B cell / target cell aggregates that show either increased or decreased Ca2 + efflux are selected.

Following the selection step, the B cells are cultured under conditions suitable for the antibody to be secreted into the medium. The antibody produced is a monoclonal antibody. Culturing may require the use of a helper cell line, such as a thymoma helper cell line (see, eg, EL4-B5, Zubler et al., 1985, J. Biol.
Immunol. 134 (6), 3662-3668). For example, in order to exclude antibodies produced against proteins expressed on the surface of cells other than the target protein, it is preferable to perform a verification step that tests the production of antibodies that specifically bind to the target. For example, CELISA, a modified enzyme-linked immunosorbent assay (ELISA) in which the coating process is performed on whole cells, is suitable for this purpose. The method allows an assessment of the selectivity and ability of antibodies that compete with the ligand.

  The antibody produced in the above step is then analyzed to identify the CDRs of the antibody. This may require steps such as antibody purification, elucidation of their amino acid sequences and / or nucleic acid sequences.

  Finally, CDRs can be fused to an acceptor framework, preferably the above-described acceptor framework, for example by gene synthesis using an oligo extension method. In one embodiment, art-recognized methods of CDR fusion can be used to transfer donor CDRs into the acceptor framework. In most cases, all three CDRs from the heavy chain are grafted from the donor antibody to a single acceptor framework, and all three CDRs from the light chain are grafted to another acceptor framework. Some CDRs may not be involved in binding to the antigen, and CDRs with different sequences (and the same length) may have the same folding (thus, despite the different sequences, the antigen It is anticipated that it will not always be necessary to implant all of these CDRs since contact from the main chain to main chain contact can be retained). Indeed, even a single domain (Ward et al., 1989, Nature, 341, 544-546) or even a single CDR (R. Taub et al., 1989, J. Biol Chem, 264, 259-265) alone. It can have antigen binding activity. However, whether all CDRs are transplanted or only some of the CDRs are transplanted, the intent of the CDR fusion is to transplant the same or nearly identical antigen binding site from an animal to a human antibody (eg, U.S. Pat. No. 5,225,539 (Winter)).

  In another embodiment, the CDRs of the donor antibody can be modified before or after being incorporated into the acceptor framework.

  Instead, antibody characterization is performed only in their final immunobinder format. For this approach, the CDR sequences of antibodies expressed on sorted B cells are obtained by RT-PCR directly from sorted B cell cultures or from a single cell of sorted B cells. to recover. For that purpose, the proliferation of B cells and / or the verification steps described above and / or the analysis steps described above may be unnecessary. A one-step PCR procedure with a combination of two partially overlapping oligonucleotides, one oligonucleotide pool encoding the CDR and the second pool encoding the framework region of a suitable immunobinder scaffold. It is possible to produce humanized immunobinders. Instead of characterizing the IgG secreted into the cell culture supernatant, high-throughput sequencing, cloning, and production allow for the performance of clonal selection based on the performance of purified humanized immunobinders. In its preferred embodiment, the immunobinder is scFv.

  However, CDR fusion can result in a loss of affinity of the generated immunobinder for the antigen due to framework residues in contact with the antigen. Such an interaction may be the result of somatic hypermutation. Therefore, fusion of such donor framework amino acids to the framework of the humanized antibody may still be necessary. The amino acid residues of the non-human immunobinder that are responsible for antigen binding can be identified by examination of the rabbit monoclonal antibody variable region sequence and structure. Each residue in the CDR donor framework that is different from the germline can be considered important. If the closest germline cannot be determined, the sequences can be compared against a subgroup consensus or a consensus of rabbit sequences with a high percentage of similarity. Rare framework residues may be the result of somatic hypermutation and are therefore thought to play a role in binding.

The antibodies of the invention can be further optimized to exhibit enhanced functional properties, such as enhanced solubility and / or stability. In certain embodiments, the antibodies of the present invention are disclosed in the “Sequence Based Engineering and Optimization of Application, filed Mar. 12, 2008, which is incorporated herein by reference.
Optimized according to the “functional consensus” method disclosed in PCT application number PCT / EP2008 / 001958, entitled “Single Chain Antibodies”.

  For example, comparing an immunobinder of the present invention with a database of functionally selected scFvs used to identify amino acid residue positions that are more or less resistant to variability than the corresponding position in an immunobinder, Thereby, it can be shown that such identified residue positions may be suitable for manipulation to improve functionality such as stability and / or solubility.

Exemplary framework residue positions for substitution and exemplary framework substitution are described in PCT application PCT application filed June 25, 2008, entitled “Methods of Modifying Antibodies, and Modified Antibodies with Improved Functional Properties”. CH2008 / 000285 and PCT application No. PCT / CH2008 / 000284 filed on June 25, 2008, entitled “Sequence Based Engineering and Optimization of Single Chain Antibodies”. For example, one or more of the following substitutions can be introduced at the amino acid position in the heavy chain variable region of the immunobinder of the present invention (see AHo numbering for each amino acid position listed below): : (A) Q or E at amino acid position 1;
(B) Q or E at amino acid position 6;
(C) T, S or A at amino acid position 7, more preferably T or A, even more preferably T;
(D) A, T, P, V, or D at amino acid position 10, more preferably T, P, V, or D,
(E) L or V at amino acid position 12, more preferably L,
(F) V, R, Q, M, or K at amino acid position 13, more preferably V, R, Q, or M;
(G) R, M, E, Q or K at amino acid position 14, more preferably R, M, E or Q, even more preferably R or E;
(H) L or V at amino acid position 19, more preferably L;
(I) R, T, K, or N at amino acid position 20, more preferably R, T, or N, even more preferably N;
(J) I, F, L, or V at amino acid position 21, more preferably I, F, or L, even more preferably I or L;
(K) R or K at amino acid position 45, more preferably K;
(L) T, P, V, A or R at amino acid position 47, more preferably T, P, V or R, even more preferably R;
(M) K, Q, H, or E at amino acid position 50, more preferably K, H, or E, even more preferably K;
(N) M or I at amino acid position 55, more preferably I;
(O) K or R at amino acid position 77, more preferably K;
(P) A, V, L, or I at amino acid position 78, more preferably A, L, or I, even more preferably A;
(Q) E, R, T, or A at amino acid position 82, more preferably E, T, or A, even more preferably E;
(R) T, S, I, or L at amino acid position 86, more preferably T, S, or L, even more preferably T;
(S) D, S, N, or G at amino acid position 87, more preferably D, N, or G, even more preferably N;
(T) A, V, L, or F at amino acid position 89, more preferably A, V, or F, even more preferably V;
(U) F, S, H, D, or Y at amino acid position 90, more preferably F, S, H, or D;
(V) D, Q, or E at amino acid position 92, more preferably D or Q, even more preferably D;
(W) G, N, T, or S at amino acid position 95, more preferably G, N, or T, even more preferably G;
(X) T, A, P, F, or S at amino acid position 98, more preferably T, A, P, or F, even more preferably F;
(Y) R, Q, V, I, M, F, or L at amino acid position 103, more preferably R, Q, I, M, F, or L, even more preferably Y, even more preferably L And (z) N, S, or A at amino acid position 107, more preferably N or S, even more preferably N.

Additionally or alternatively, one or more of the following substitutions can be introduced into the light chain variable region of the immunobinder of the invention:
(Aa) Q, D, L, E, S, or I at amino acid position 1, more preferably L, E, S, or I, even more preferably L or E;
(Bb) S, A, Y, I, P, or T at amino acid position 2, more preferably A, Y, I, P, or T, even more preferably P or T;
(Cc) Q, V, T, or I at amino acid position 3, more preferably V, T, or I, even more preferably V or T;
(Dd) V, L, I, or M at amino acid position 4, more preferably V or L;
(Ee) S, E, or P at amino acid position 7, more preferably S or E, even more preferably S;
(Ff) T or I at amino acid position 10, more preferably I;
(Gg) A or V at amino acid position 11, more preferably A;
(Hh) S or Y at amino acid position 12, more preferably Y;
(Ii) T, S, or A at amino acid position 14, more preferably T or S, even more preferably T;
(Jj) S or R at amino acid position 18, more preferably S;
(Kk) T or R at amino acid position 20, more preferably R;
(Ll) R or Q at amino acid position 24, more preferably Q;
(Mm) H or Q at amino acid position 46, more preferably H;
(Nn) K, R, or I at amino acid position 47, more preferably R or I, even more preferably R;
(Oo) R, Q, K, E, T or M at amino acid position 50, more preferably Q, K, E, T or M;
(Pp) K, T, S, N, Q, or P at amino acid position 53, more preferably T, S, N, Q, or P;
(Qq) I or M at amino acid position 56, more preferably M;
(Rr) H, S, F, or Y at amino acid position 57, more preferably H, S, or F;
(Ss) I, V, or T at amino acid position 74, more preferably V or T, R, even more preferably T;
(Tt) R, Q, or K at amino acid position 82, more preferably R or Q, even more preferably R;
(Uu) L or F at amino acid position 91, more preferably F;
(Vv) G, D, T, or A at amino acid position 92, more preferably G, D, or T, even more preferably T;
(Xx) S or N at amino acid position 94, more preferably N;
(Yy) F, Y, or S at amino acid position 101, more preferably Y or S, even more preferably S; and (zz) D, F, H, E, L, A, T at amino acid position 103. , V, S, G, or I, more preferably H, E, L, A, T, V, S, G, or I, even more preferably A or V.

  In other embodiments, the immunobinders of the present invention have stability described in US Provisional Application No. 61 / 075,692, filed June 25, 2008, entitled “Solubility Optimization of Immunobinders”. One or more mutations that enhance In certain preferred embodiments, the immunobinder comprises a mutation that enhances solubility at an amino acid position selected from the group of heavy chain amino acid positions consisting of 12, 103, and 144 (AHo numbering method). In a preferred embodiment, the immunobinder comprises (a) a serine (S) at heavy chain amino acid position 12; (b) a serine (S) or threonine (T) at heavy chain amino acid position 103; and (c) a heavy chain amino acid. It includes one or more substitutions selected from the group consisting of serine (S) or threonine (T) at position 144. In another embodiment, the immunobinder has the following substitutions: (a) serine (S) at heavy chain amino acid position 12; (b) serine (S) or threonine (T) at heavy chain amino acid position 103; c) Contains serine (S) or threonine (T) at heavy chain amino acid position 144.

  In certain preferred embodiments, the immunobinder is a light chain acceptor framework at at least one of positions 1, 3, 4, 10, 47, 57, 91 and 103 of the light chain variable region according to the AHo numbering system. Contains stability enhancing mutations at the framework residues. In a preferred embodiment, the light chain acceptor framework comprises (a) glutamic acid at position 1 (E), (b) valine at position 3 (V), (c) leucine at position 4 (L), (d) position. 10 serine (S), (e) arginine (R) at position 47, (e) serine (S) at position 57, (f) phenylalanine (F) at position 91, and (g) valine at position 103 (V ) Comprising one or more substitutions selected from the group consisting of:

  Any of a variety of available methods for producing humanized antibodies containing the mutations described above can be used.

  Accordingly, the present invention provides immunobinders that are humanized according to the methods described herein.

  In certain preferred embodiments, the immunobinder target antigen is VEGF or TNFα.

  For example, the polypeptides described in the present invention or the polypeptides produced by the methods of the present invention can be synthesized using techniques known in the art. Alternatively, a polypeptide can be produced by synthesizing a nucleic acid molecule encoding the desired variable region and by recombinant methods.

  For example, once the sequence of a humanized variable region is determined, the variable region or a polypeptide containing it can be generated by techniques well known in the field of molecular biology. More specifically, recombinant DNA techniques can be used to produce a wide range of polypeptides by transforming host cells with nucleic acid sequences (eg, DNA sequences encoding the desired variable region ( For example, modified heavy or light chain; variable domains or other antigen-binding fragments thereof)).

In one embodiment, an expression vector can be prepared that includes a promoter operably linked to a DNA sequence encoding at least _VH or _VL . If necessary or desirable, a second expression vector can be prepared that contains a promoter operably linked to the DNA sequence encoding the complementary variable domain (ie, the parent expression vector encodes V _H , When two expression vectors encode _VL and vice versa). A cell line (eg, an immortalized mammalian cell line) can then be transformed with one or both of the expression vectors and cultured under conditions that allow expression of the chimeric variable domain or chimeric antibody (eg, (See Neuberger et al., International Patent Application No. PCT / GB85 / 00392).

  In one embodiment, a variable region comprising the amino acid sequences of the donor CDR and acceptor FR can be created and then introduced into the nucleic acid molecule to result in a CDR amino acid substitution.

  Exemplary methods recognized in the art for making nucleic acid molecules encoding amino acid sequence variants of a polypeptide include site-specific (or oligonucleotide-mediated) mutagenesis, PCR mutagenesis, and the polypeptide. This includes, but is not limited to, preparation of encoded pre-prepared DNA by cassette mutagenesis.

  Site-directed mutagenesis is a preferred method for preparing substitutional variants. This technique is well known in the art (eg, Carter et al., Nucleic Acids Res., 13, 4431-4443 (1985), Kunkel et al., Proc. Natl. Acad. Sci. USA, 82, 488). Volume (1987)). Briefly, when performing site-directed mutagenesis of DNA, modification of the parent DNA is performed by first hybridizing an oligonucleotide encoding the desired mutation to a single strand of the parent DNA. Following hybridization, the entire second strand is synthesized using DNA polymerase using the hybridized oligonucleotide as a primer and the single strand of the parent DNA as a template. Thereby, an oligonucleotide encoding the desired mutation is incorporated into the resulting double-stranded DNA.

  PCR mutagenesis is also suitable for making amino acid sequence variants of polypeptides. Higuchi, PCR Protocols, 177-183 (Academic Press, 1990); Vallette et al., Nuc. Acids Res. 17, 723-733 (1989). Briefly, when a small amount of template DNA is used as a starting material in PCR, a primer that is slightly different in sequence from the corresponding region of the template DNA is used and differs from the template sequence only in the position where the primer differs from the template. A relatively large amount of specific DNA fragments can be produced.

  Another method for preparing variants, cassette mutagenesis, is based on the technique described by Wells et al., Gene, 34, 315-323 (1985). The starting material is a plasmid (or other vector) containing the DNA to be mutated. Identify the codon (s) in the parent DNA to be mutated. There must be a unique restriction endonuclease site on each side of the identified mutation site (s). If no such restriction enzyme recognition site is present, it can be generated using the above-described oligonucleotide-mediated mutagenesis method that introduces a restriction enzyme recognition site at an appropriate location in the DNA encoding the polypeptide. . The plasmid DNA is cut at these sites to linearize it. A double-stranded oligonucleotide that encodes the sequence of DNA between the restriction enzyme recognition sites but contains the desired mutation (s) is synthesized using standard procedures, where the two strands of the oligonucleotide Are synthesized separately using standard techniques and then hybridized to each other. This double-stranded oligonucleotide is called a cassette. This cassette is designed to have 5 'and 3' ends that are compatible with the ends of the linearized plasmid, such that it is directly ligated to the plasmid. This plasmid now contains the mutated DNA sequence.

  The variable region generated by the method of the present invention can be remodeled and further modified to further enhance antigen binding. Thus, the above steps can precede or follow additional steps including, for example, affinity maturation. In addition, empirical combined data can be used for further optimization.

  One skilled in the art will appreciate that the polypeptides of the invention can be further modified to differ in amino acid sequence but not the desired activity. For example, additional nucleotide substitutions may be added to the protein that lead to amino acid substitutions at “non-essential” amino acid residues. For example, a non-essential amino acid residue in an immunoglobulin polypeptide can be replaced with another amino acid residue from the same side chain family. In another embodiment, a sequence of amino acids can be replaced with a structurally similar sequence that differs in order and / or composition of side chain family members, ie, the amino acid residues have similar side chains Conservative substitutions can be made that are substituted with amino acid residues.

  Families of amino acid residues having similar side chains have been defined in the art and include basic side chains (eg, lysine, arginine, histidine), acidic side chains (eg, aspartic acid, glutamic acid), Uncharged polar side chains (eg glycine, asparagine, glutamine, serine, threonine, tyrosine, cysteine), nonpolar side chains (eg alanine, valine, leucine, isoleucine, proline, phenylalanine, methionine, tryptophan), beta branch Side chains (eg threonine, valine, isoleucine) and aromatic side chains (eg tyrosine, phenylalanine, tryptophan, histidine).

  Apart from amino acid substitutions, the present invention also contemplates other modifications to the Fc region amino acid sequence, eg, to generate Fc region variants with altered effector functions. For example, one or more amino acid residues of the Fc region can be removed to reduce or facilitate binding to FcR. In one embodiment, one or more Fc region residues can be modified to generate such Fc region variants. Usually, according to this embodiment of the invention, no more than 1 to about 10 Fc region residues are removed. Wherein the Fc region comprising one or more amino acid deletions preferably retains at least about 80%, preferably at least about 90%, most preferably at least about 95% of the starting Fc region or native sequence human Fc region. To do.

  Amino acid inserted Fc region variants with altered effector functions can also be generated. For example, at least one amino acid residue adjacent to one or more Fc region positions identified herein as impacting FcR binding (eg, 1 to 2 amino acid residues, usually no more than 10 residues) ) Can be introduced. “Contiguous” means within 1-2 amino acid residues of the Fc region residues identified herein. Such Fc region variants may exhibit enhanced or reduced FcRn binding.

  Such Fc region variants typically comprise at least one amino acid modification within the Fc region. In one embodiment, amino acid modifications can be combined. For example, a variant Fc region can include substitutions therein, such as 2, 3, 4, 5, etc., of the specific Fc region positions identified herein. In another embodiment, the polypeptide may have an altered binding to FcRn and binding to another Fc receptor.

  In one embodiment, a polypeptide described in the present invention or a polypeptide produced by a method of the present invention, eg, a humanized Ig variable region and / or a polypeptide comprising a humanized Ig variable region, is recombinant. It may have been produced by the law. For example, a polynucleotide sequence encoding a polypeptide can be inserted into an expression vector suitable for recombinant expression. When the polypeptide is an antibody, additional light chain variable region and polynucleotide encoding heavy chain variable regions, optionally linked to the constant region, can be inserted into the same or different expression vectors. An affinity tag sequence (eg, a His (6) tag) can optionally be linked or included in the polypeptide sequence to facilitate downstream purification. The DNA segment encoding the immunoglobulin chain is operably linked to control sequences in the expression vector (s) that ensure expression of the immunoglobulin polypeptide. Expression control sequences include, but are not limited to, promoters (eg, naturally associated or heterologous promoters), signal sequences, enhancer elements and transcription termination sequences. The expression control sequence is preferably a eukaryotic promoter system in a vector capable of transforming or transfecting a eukaryotic host cell. Once the vector is incorporated into a suitable host, the host is maintained under conditions suitable for high level expression of the nucleotide sequence and collection and purification of the polypeptide.

  These expression vectors are usually replicable in the host organism as an episome or as a part integrated into the host chromosomal DNA. In general, expression vectors contain a selectable marker (eg, ampicillin resistance, hygromycin resistance, tetracycline resistance or neomycin resistance) to allow detection of cells transformed with the desired DNA sequence. (See, eg, US Pat. No. 4,704,362).

  E. E. coli is one prokaryotic host particularly useful for cloning the polynucleotides (eg, DNA sequences) of the present invention. Other microbial hosts suitable for use include Neisseria gonorrhoeae such as Bacillus subtilus, as well as other Enterobacteriaceae such as Salmonella, Serratia and various Pseudomonas species.

  Other microorganisms such as yeast are also useful for expression. Saccharomyces and Pichia are exemplary yeast hosts, and suitable vectors optionally have expression control sequences (eg, promoters), origins of replication, termination sequences, and the like. Typical promoters include 3-phosphoglycerate kinase and other glycolytic enzymes. Inducible yeast promoters include, inter alia, promoters derived from alcohol dehydrogenase and isocytochrome C, and promoters derived from enzymes responsible for the use of methanol, maltose and galactose.

  Within the scope of the present invention, E.I. coli and S. coli. cerevisiae is a preferred host cell.

In addition to microorganisms, mammalian tissue culture can also be used to express and produce the polypeptides of the invention (eg, polynucleotides encoding immunoglobulins or fragments thereof). Winnacker, From Genes to Clones, VCH
Publishers, N.M. Y. , N.M. See Y (1987). Since many suitable host cell systems capable of secreting heterologous proteins (eg, intact immunoglobulins) have been developed in the art, eukaryotic cells are in fact preferred, including CHO cell lines, various Cos cells Lines, HeLa cells, 293 cells, myeloma cell lines, transformed B cells and hybridomas. Expression vectors for these cells include replication origins, expression control sequences such as promoters and enhancers (Queen et al., Immunol. Rev., 89, 49 (1986)), as well as ribosome binding sites, RNA splicing sites. Necessary processing information sites such as polyadenylation sites and transcription terminator sequences can be included. Preferred expression control sequences are promoters obtained from immunoglobulin genes, SV40, adenovirus, bovine papilloma virus, cytomegalovirus and the like. Co et al. Immunol. 148, 1149 (1992).

  Vectors containing the polynucleotide sequences of interest (eg, sequences encoding heavy and light chains and expression control sequences) can be introduced into host cells by well-known methods (depending on the type of cell host). For example, calcium chloride transfection is commonly utilized for prokaryotic cells, while calcium phosphate treatment, electroporation, lipofection, biolistics or virus-based transfection can be used for other cellular hosts (general Sambrook et al., Molecular Cloning: A Laboratory Manual (see Cold Spring Harbor Press, 2nd edition, 1989) Other methods used to transform mammalian cells include the use of polybrene, protoplast fusion , Liposomes, electroporation, and microinjection (see generally Sambrook et al., Ibid.) For the production of transgenic animals , That microinjected transgene into fertilized oocytes, or incorporated into the genome of embryonic stem cells, it can be introduced the nuclei of such cells into enucleated intracellularly.

  The polypeptide of interest can also be incorporated into a transgene for introduction into the genome of the transgenic animal and subsequent expression, for example, in the milk of the transgenic animal (eg, Deboer et al., 5, 741,957; Rosen, 5,304,489; and Meade, 5,849,992). Suitable transgenes include light and / or heavy chain coding sequences operably linked to promoters and enhancers from mammary gland specific genes such as casein or beta-lactoglobulin.

  Polypeptides can be expressed using a single vector or can be expressed using two vectors. For example, antibody heavy and light chains may be cloned into separate expression vectors and co-transfected into cells.

  In one embodiment, a signal sequence can be used to facilitate expression of a polypeptide of the invention.

  Once expressed, the polypeptide can be clarified using standard procedures in the art (general, including ammonium sulfate precipitation, affinity columns (eg, Protein A or Protein G), column chromatography, HPLC purification, gel electrophoresis, etc.) In particular, see Scopes, Protein Purification (see Springer-Verlag, NY (1982))).

  Humanized Ig variable regions or polypeptides comprising them can be expressed by host cells or cultured cell lines. They can also be expressed in cells in vivo. The cell line that is transformed (eg, transfected) to produce the modified antibody can be an immortalized mammalian cell line, such as that of a lymphoid origin (eg, myeloma, hybridoma). , Trioma or quadroma cells). This cell line can also include normal lymphoid cells such as B cells that have been immortalized by transformation with a virus (eg, Epstein-Barr virus).

  The cell line used to produce the polypeptide is usually a mammalian cell line, but cell lines from other sources (such as bacteria and yeast) can also be used. In particular, E.I. E. coli-derived bacterial strains can be used, among others, for example, phage display.

  Some immortal lymphoid cell lines, such as myeloma cell lines, secrete isolated Ig light or heavy chains in their normal state. When such a cell line is transformed with a vector expressing a modified antibody prepared during the method of the present invention, the normally secreted chain is a variable domain of an Ig chain encoded by a previously prepared vector. The remaining steps of the method need not be performed provided that the strands are complementary to each other.

  If the immortalized cell line does not secrete or does not secrete the complementary strand, it is necessary to introduce a vector encoding the appropriate complementary strand or fragment thereof into the cell.

  If the immortal cell line secretes a complementary light or heavy chain, the transformed cell line can, for example, transform a suitable bacterial cell with the vector and then transform the bacterial cell with the immortal cell line (eg, , By spheroplast fusion). Alternatively, DNA can be introduced directly into an immortalized cell line by electroporation.

In one embodiment, the humanized Ig variable region described in the present invention, or the humanized Ig variable region produced by the method of the present invention, can be present in any antigen-binding fragment of an antibody. These fragments can be produced recombinantly, as well as engineered, synthesized or produced by digesting the antibody with proteolytic enzymes. For example, the fragment can be a Fab fragment, and digestion with papain cleaves the antibody at the region preceding the interchain (ie, V _H -V _H ) disulfide bond connecting the two heavy chains. This results in the formation of two identical fragments containing the V _H and C _H 1 domains of the light chain and the heavy chain. Alternatively, this fragment can be an F (ab ′) ₂ fragment. These fragments can be generated by digesting the antibody with pepsin. Pepsin cleaves the heavy chain after an interchain disulfide bond, resulting in a fragment containing both antigen binding sites. Yet another alternative is to use “single chain” antibodies. Single chain Fv (scFv) fragments can be constructed in a variety of ways. For example, connecting the C-terminus of _{V H} to the N-terminus of the _{V L.} Usually, a linker (eg (GGGGS) ₄ ) is placed between V _H and V _L. However, the order in which the strands can be ligated can be reversed and can include tags (eg, Myc tag, His tag or FLAG tag) that facilitate detection or purification (such tags as those of the present invention). Can be added to any antibody or antibody fragment; their use is not limited to scFv). Thus, as described below, tagged antibodies are within the scope of the present invention. In an alternative embodiment, the antibodies described herein or the antibodies produced by the methods described herein can be heavy chain dimers or light chain dimers. In addition, antibody light or heavy chains or portions thereof, such as single domain antibodies (DAb) can be used.

In another embodiment, the humanized Ig variable region described in the present invention or the humanized Ig variable region generated by the method of the present invention is present in a single chain antibody (ScFv) or minibody (minibody) ( For example, see US Pat. No. 5,837,821 or WO94 / 09817A1). Each minibody is a ScFv molecule (a single polypeptide containing one or more antigen binding sites, eg, a _VL domain linked by a flexible linker to a _VH domain fused to a CH3 domain via a connecting peptide). Is a dimer molecule made up of two polypeptide chains. ScFv molecules can be constructed in a _VH -linker- _VL orientation or a _VL -linker- _VH orientation. The flexible hinge that connects the _VL and _VH domains that make up the antigen binding site preferably comprises from about 10 to about 50 amino acid residues. An exemplary connecting peptide for this purpose is (Gly4Ser) 3 (Huston et al. (1988), PNAS, 85, 5879). Other connecting peptides are known in the art.

Methods for producing single chain antibodies are well known in the art, for example, Ho et al. (1989), Gene, 77, 51; Bird et al. (1988) Science, 242: 423, Pantoliano (1991), Biochemistry, 30, 10117; Milenic et al. (1991), Cancer Research, 51, 6363; Takkinen et al. (1991), Protein Engineering, 4, 837. Minibodies can be generated by constructing a ScFv component and a connecting peptide-CH ₃ component using methods described in the art (see, eg, US Pat. No. 5,837,821 or WO94 / 09817A1). . These components can be isolated as restriction fragments from separate plasmids and then ligated and recloned into an appropriate vector. Appropriate assembly can be verified by restriction enzyme digestion and DNA sequence analysis. In one embodiment, the minibody of the invention comprises a connecting peptide. In one embodiment, the connecting peptide comprises a Gly / Ser linker, such as GGGSSSGGGSGG.

In another embodiment, a tetravalent minibody can be constructed. For example, a tetravalent minibody can be constructed in the same way as a minibody except that two ScFv molecules are linked using a flexible linker having the amino acid sequence (G ₄ S) ₄ G ₃ AS. it can.

In another embodiment, the humanized variable region described in the present invention, or the humanized variable region produced by the method of the present invention, can be present in a diabody. Diabodies are similar to scFv molecules, but usually have a short (less than 10, preferably 1-5) amino acid residue linker connecting both variable domains, and thus V on the same polypeptide chain. _L domain and V _H domain cannot interact. Instead, the V _L and V _H regions of one polypeptide chain (respectively) interact with the V _H and V _L domains on the second polypeptide chain (WO 02/02781).

  In another embodiment, a humanized variable region of the invention comprises an immunoreactive fragment of an antibody (eg, scFv molecule, minibody, tetravalent minibody, or diabody) operably linked to an FcR binding moiety or Can exist in the part. In one exemplary embodiment, the FcR binding portion is a complete Fc region.

  The humanization methods described herein preferably result in an Ig variable region that has substantially no change in antigen affinity compared to the donor antibody.

In one embodiment, a polypeptide comprising a variable domain of the invention has about 10 ⁵ M ⁻¹ , about 10 ⁶ M ⁻¹ , about 10 ⁷ M ⁻¹ , about 10 ⁸ M ⁻¹ , about 10 ⁹ M ^−1. Binds to an antigen with a binding affinity Ka of a binding constant Ka greater than (or equal to) about 10 ¹⁰ M ⁻¹ , about 10 ¹¹ M ⁻¹ or about 10 ¹² M ⁻¹ (intermediate affinity of these values) including).

  Affinity, avidity and / or specificity can be measured in various ways. Usually, regardless of the exact method by which affinity is defined or measured, the method of the invention produces an antibody that is superior to the original antibody (or antibodies) in any aspect of clinical application. The method improves antibody affinity (eg, if the modified antibody can be administered at a lower dose or less frequently or by a convenient route of administration than the antibody (or antibodies) of origin, The method of the present invention is considered to be effective or successful.

More particularly, the affinity between an antibody and the antigen to which it binds can vary, including, for example, an ELISA assay, BiaCore assay, or KinExA ^™ 3000 assay (available from Sapidyne Instruments (Boise (ID))). Can be measured with a simple assay. Briefly, antigens (eg, antigens used in the methods of the invention may be any antigen of interest (eg, cancer antigen; cell surface protein or secreted protein); pathogen antigen (eg, bacterial or Coating with a viral antigen (eg, HIV antigen, influenza antigen or hepatitis antigen); which may be an allergen) Prepare dilutions of the antibody to be tested and add each dilution to the designated hole in the plate. A detection antibody (eg, goat anti-human IgG-HRP conjugate) is then added to each well, followed by a chromogenic substrate (eg, HRP) The plate is then loaded at 450 nM with an ELISA plate reader. Read and calculate EC50 values (the methods described here are generally applicable; Comprising particular it is understood as not being limited to the production of antigens or antibodies that bind to the class of antigens).

  Those skilled in the art recognize that determining affinity is not as simple as always looking at a single number. Since antibodies have two arms, their apparent affinity is usually much higher than the intrinsic affinity between the variable region and the antigen (this is believed to be due to avidity). Intrinsic affinity can be measured using scFv or Fab fragments.

  In another aspect, the invention features a bispecific molecule comprising a humanized rabbit antibody or fragment thereof of the invention. An antibody of the invention, or antigen-binding portion thereof, is derivatized or linked to another functional molecule such as, for example, another peptide or protein (eg, another antibody or a ligand for a receptor) to provide at least 2 Bispecific molecules can be produced that bind to two different binding sites or target molecules. The antibodies of the invention can be derivatized or linked to two or more other functional molecules to produce multispecific molecules that bind to three or more different binding sites and / or target molecules. Such multispecific molecules can also be intended to be encompassed by the term “bispecific molecule” as used herein. To create a bispecific molecule of the present invention, an antibody of the present invention can be combined with another antibody, antibody fragment, tumor-specific or pathogen-specific antigen, peptide, or one such as a binding mimetic. Or, it may be operably linked (eg, by chemical coupling, gene fusion, non-covalent association, or the like) to a plurality of other binding molecules such that a bispecific molecule results. Accordingly, the present invention provides a bispecific comprising at least one first binding molecule having specificity for a first target and a second binding molecule having specificity for one or more additional target epitopes. Contains sex molecules.

  In one embodiment, the bispecific molecule of the invention has at least one antibody, or antibody fragment thereof (eg, Fab, Fab ′, F (ab ′) 2, Fv, or single chain Fv) as binding specificity. Including). An antibody can be a light chain or heavy chain dimer, or any minimal fragment thereof (as described in Ladner et al., US Pat. No. 4,946,778, the contents of which are expressly incorporated by reference). Fv or single chain constructs etc.).

  Although human monoclonal antibodies are preferred, other antibodies that can be used in the bispecific molecules of the invention are murine, chimeric, and humanized monoclonal antibodies.

  The bispecific molecules of the present invention can be prepared by conjugating the binding specificity of the constituents using methods known in the art. For example, each binding specificity of the bispecific molecule may be produced separately and then conjugated to each other. If the binding specificity is a protein or peptide, various coupling agents or cross-linking agents can be used for covalent conjugation. Examples of the cross-linking agent include protein A, carbodiimide, N-succinimidyl-S-acetyl-thioacetate (SATA), 5,5′-dithiobis (2-nitrobenzoic acid) (DNTB), o-phenylene dimaleimide ( oPDM), N-succinimidyl-3- (2-pyridyldithio) propionate (SPDP), and sulfosuccinimidyl 4- (N-maleimidomethyl) cyclohexane-1-carboxylate (sulfo-SMCC) (e.g. Karpovsky et al. (1984) J. Exp. Med. 160, 1686; Liu, MA et al. (1985) Proc. Natl. Acad. Sci. USA, 82, 8648). Other methods include Paulus (1985) Behring Ins. Mitt. 78, 118-132; Brennan et al. (1985) Science 229, 81-83, and Glennie et al. (1987) J. MoI. Immunol. 139, 2367-2375 pages. Preferred conjugating agents are SATA and sulfo-SMCC, both of which are Pierce Chemical Co. (Rockford, IL).

  If the binding specificity is an antibody, they may be conjugated by sulfhydryl binding of the C-terminal hinge region of the two heavy chains. In one particularly preferred embodiment, the hinge region is modified to include an odd number, preferably one sulfhydryl residue, and then conjugated.

Alternatively, both binding specificities may be encoded in the same vector and expressed and assembled in the same host cell. This method is particularly useful when the bispecific molecule is a mAb × mAb, mAb × Fab, Fab × F (ab ′) ₂ , or ligand × Fab fusion protein. The bispecific molecule of the invention may be a single chain molecule comprising one single chain antibody and a binding determinant, or a single chain bispecific molecule comprising two binding determinants. Bispecific molecules may include at least two single chain molecules. Methods for preparing bispecific molecules are described, for example, in US Pat. No. 5,260,203, US Pat. No. 5,455,030, US Pat. No. 4,881,175, US Pat. 132,405, US Pat. No. 5,091,513, US Pat. No. 5,476,786, US Pat. No. 5,013,653, US Pat. No. 5,258,498, and US Pat. , 482,858.

Binding to their specific targets by bispecific molecules can be achieved, for example, by enzyme-linked immunosorbent assay (ELISA), radioimmunoassay (RIA), flow cytometry-based single cell sorting (eg, FACS analysis), bioassay (Eg, growth inhibition) or Western blot assay. Each of these assays generally detects the presence of the protein-antibody complex of interest, in particular by using a labeled reagent (eg, an antibody) specific for the complex of interest. For example, an antibody complex can be detected using, for example, an enzyme-linked antibody or antibody fragment that recognizes and specifically binds to an antibody-VEGF complex. Alternatively, the complex can be detected using a variety of any other immunoassay. For example, the antibody may be radiolabeled and used in radioimmunoassay (RIA) (eg, Weintraub, B., Principles of Radioimmunoassays, Seventh Training Course on, incorporated herein by reference).
(See Radioligand Assay Technologies, The Endocrine Society, March 1986). The radioactive isotope can be detected by such means as the use of a gamma counter or a scintillation counter or by autoradiography.

  In another aspect, the invention features a humanized rabbit antibody, or fragment thereof, conjugated to a therapeutic moiety, such as a cytotoxin, drug (eg, immunosuppressive drug), or radiotoxin. Such a complex is referred to herein as an “immunoconjugate”. An immunoconjugate containing one or more cytotoxins is called an “immunotoxin”. A cytotoxin or cytotoxic agent includes any agent that is detrimental to (eg, kills) cells. Examples include taxol, cytochalasin B, gramicidin D, ethidium bromide, emetine, mitomycin, etoposide, tenoposide, vincristine, vinblastine, colchicine, doxorubicin, daunorubicin, dihydroxyanthracindione, mitoxantrone, mitromycin, actinomycin D, 1-dehydrotestosterone, glucocorticoid, procaine, tetracaine, lidocaine, propranolol, and puromycin, and analogs or homologues thereof. Examples of therapeutic agents include antimetabolites (eg, methotrexate, 6-mercaptopurine, 6-thioguanine, cytarabine, 5-fluorouracil, dacarbazine), alkylating agents (eg, mechloretamine, thiotepa, chlorambucil) , Melphalan, carmustine (BSNU), and lomustine (CCNU), cyclophosphamide, busulfan, dibromomannitol, streptozotocin, mitomycin C, and cis-dichlorodiamineplatinum (II) (DDP), cisplatin, anthra Cyclins (eg, daunorubicin (formerly daunomycin) and doxorubicin), antibiotics (eg, dactinomycin (formerly activins) Mycin), bleomycin, mithramycin, and anthramycin (AMC)), and anti-mitotic agents (e.g., vincristine and vinblastine) can be cited also.

Other preferred examples of therapeutic cytotoxins that can be conjugated to the antibodies of the invention include duocarmycin, calicheamicin, maytansine, and auristatin, and derivatives thereof. An example of a calicheamicin antibody conjugate is commercially available (Mylotarg ^™ , Wyeth-Ayerst).

  Cytotoxins can be conjugated to the antibodies of the present invention using linker techniques available in the art. Examples of types of linkers used to conjugate cytotoxins to antibodies include, but are not limited to, hydrazones, thioethers, esters, disulfides, and peptide-containing linkers. Linkers are susceptible to cleavage by proteases, such as those that are susceptible to cleavage by low pH within the lysosomal compartment, or proteases that are preferentially expressed in tumor tissue, such as cathepsins (eg, cathepsins B, C, D). Easy ones can be selected.

  For further discussion of cytotoxins, linker types, and methods for conjugating therapeutic agents to antibodies, Saito, G. et al. Et al. (2003) Adv. Drug Deliv. Rev. 55, 199-215; Trail, P.M. A. Et al. (2003) Cancer immunol. Immunother. 52, 328-337; Payne, G .; (2003) Cancer Cell, 3, 207-212; Allen, T .; M.M. (2002) Nat. Rev. Cancer Vol. 2, 750-763; Pastan, I .; And Kreitman, R .; J. et al. (2002) Curr. Opin. Investig. Drugs, 3, 1089-1091; D. And Springer, C.I. J. et al. (2001) Adv. Drug Deliv. Rev. 53, pp. 247-264.

The antibodies of the invention can also be conjugated to a radioisotope to produce a cytotoxic radiopharmaceutical, also called a radioimmunoconjugate. Examples of radioisotopes that can be conjugated to antibodies for use in diagnosis or therapy include, but are not limited to, iodine ¹³¹ , indium ¹¹¹ , yttrium ⁹⁰ , and lutetium ¹⁷⁷ . Methods for preparing radioimmunoconjugates are established in the art. Examples of radioimmunoconjugates are commercially available, including Zevalin ^™ (IDEC Pharmaceuticals) and Bexxar ^™ (Corixa Pharmaceuticals), using similar methods to prepare radioimmunoconjugates using the antibodies of the invention. Can do. The antibody conjugates of the invention can be used to modulate a given biological response and the drug moiety should not be construed as limited to classical chemotherapeutic agents. For example, the drug moiety can be a protein or polypeptide having the desired biological activity. Such proteins include, for example, enzymatically active toxins, or active fragments thereof, such as abrin, ricin A, Pseudomonas exotoxins, or diphtheria toxins; proteins such as tumor necrosis factor or interferon-γ; or Biological response modifiers such as lymphokines, interleukin-1 (“IL-1”), interleukin-2 (“IL-2”), interleukin-6 (“IL-6”), granulocyte macrophages Colony stimulating factor (“GM-CSF”), granulocyte colony stimulating factor (“G-CSF”), or other growth factors may be mentioned.

Techniques for conjugating such therapeutic moieties to antibodies are well known, for example, in Monoclonal Antibodies And Cancer Therapies, Reisfeld et al. (Edit), pages 243-56, (Alan R. Liss, Inc. 1985). , Arnon et al., "Monoclonal Antibodies for Immunotargeting Of Drugs In Cancer Therapy"; Controlled Drug Delivery (2nd edition), rev. “Antibodies For Drug Delivery”; Monoclonal Antibo ies '84:
Biological And Clinical Applications, Pinchera et al. (Edited), pp. 475-506 (1985), Thorpe, “Antibody Carriers Of Cytotoxic Agents In”.
Cancer Therapy: A Review "; Monoclonal Antibodies For Cancer Detection And Therapy, Baldwin et al., (Eds.), At pp. 303~16 (Academic Press 1985 years)," Analysis, Results, And Future Prospective Of The Therapeutic Use Of Radiolabeled Antibody In Cancer Therapeutics ", and Thorpe et al.," The Preparation And Cytotoxic. "
“Properties of Antibodies-Toxin Conjugates”, Immunol. Rev. 62, 119-58 (1982).

  In one aspect, the present invention provides a pharmaceutical formulation comprising a humanized rabbit antibody for the treatment of a disease. The term “pharmaceutical formulation” is a form that allows the biological activity of an antibody or antibody derivative that is clearly effective and does not contain additional components that are toxic to the subject to which the formulation is administered. Say things. “Pharmaceutically acceptable” excipients (vehicles, additives) are those that can be reasonably administered to a subject mammal to achieve an effective dose of the active ingredient employed.

  A “stable” formulation is one in which the antibody or antibody derivative in the formulation inherently retains its physical and / or chemical stability and / or biological activity upon storage. Various analytical techniques for measuring protein stability are available in the art, see, eg, Peptide and Protein Drug Delivery, pages 247-301, edited by Vincent Lee, Marcel Dekker, Inc. New York, N .; Y. , Pubs. (1991), and Jones, A.M. Adv. Drug Delivery Rev. 10: 29-90 (1993). Stability can be measured at a selected temperature for a selected time period. Preferably, the formulation is stable at room temperature (about 30 ° C.) or 40 ° C. for at least 1 month and / or stable at about 2-8 ° C. for at least 1 year or at least 2 years. Moreover, the formulation is preferably stable after freezing (eg, up to -70 ° C.) and thawing of the formulation.

  An antibody or antibody derivative is pharmaceutical if it shows no signs of aggregation, precipitation and / or denaturation as measured by visual inspection of color and / or clarity, or by UV light scattering or size exclusion chromatography. “Retains its physical stability” in the formulation.

  An antibody or antibody derivative is said to be “in its chemical formulation” if its chemical stability at a given time is such that the protein still retains its biological activity as defined below. Maintain stability. " Chemical stability can be assessed by detecting and quantifying chemically altered protein forms. Chemical changes can be assessed using, for example, size exclusion chromatography, SDS-PAGE and / or matrix-assisted laser desorption / ionization time-of-flight mass spectrometry (MALDI / TOF MS), resizing (eg, clipping) Can be involved. Other types of chemical changes include charge changes that can be assessed by, for example, ion exchange chromatography (eg, caused by deamidation).

  The antibody or antibody derivative is within about 10% of the biological activity exhibited by the antibody's biological activity at a given time, eg, when the pharmaceutical formulation is prepared as determined in an antigen binding assay (assay Within its error) “retains its biological activity” in the pharmaceutical formulation. Other “biological activity” assays for antibodies are described in detail herein below.

  “Isotonic” means that the intended formulation has essentially the same osmotic pressure as human blood. Isotonic formulations generally have an osmotic pressure of about 250 to 350 mOsm. Isotonicity can be measured, for example, using a vapor pressure or ice freezing osmometer.

  “Polyol” is a substance having a plurality of hydroxyl groups, and includes sugars (reducing and non-reducing sugars), sugar alcohols and sugar acids. Preferred polyols herein have a molecular weight of less than about 600 kD (eg, in the range of about 120 to about 400 kD). A “reducing sugar” is one that contains a hemiacetal group that can reduce metal ions or can covalently react with lysine and other amino groups in proteins, and “non-reducing sugar” It does not have these characteristics. Examples of reducing sugars are fructose, mannose, maltose, lactose, arabinose, xylose, ribose, rhamnose, galactose and glucose. Non-reducing sugars include sucrose, trehalose, sorbose, melezitose and raffinose. Mannitol, xylitol, erythritol, threitol, sorbitol and glycerol are examples of sugar alcohols. For sugar acids, this includes L-gluconate and its metal salts. Where it is desired that the formulation be stable to freeze-thaw, the polyol preferably does not crystallize at a freezing temperature (eg, -20 ° C.) such that the antibody in the formulation becomes unstable. It is. Non-reducing sugars such as sucrose and trehalose are preferred polyols herein, and trehalose is preferred over sucrose because of the excellent solution stability of trehalose.

  As used herein, “buffer solution” refers to a buffer solution that resists changes in pH due to the action of an acid-base conjugate component. The buffer of the present invention has a pH in the range of about 4.5 to about 6.0, preferably about 4.8 to about 5.5, and most preferably about 5.0. Examples of buffers that control pH in this range include acetate (eg, sodium acetate), succinate (eg, sodium succinate), gluconate, histidine, citrate and other organic acid buffers. Can be mentioned. If a formulation that is stable to freeze-thaw is desired, the buffer is preferably not phosphate.

  In the context of the present invention, in the context of the present invention, a “therapeutically effective amount” of an antibody or antibody derivative refers to an amount that is effective in preventing or treating a disorder with respect to a treatment in which the antibody or antibody derivative is effective. A “disease / disorder” is any condition that would benefit from treatment with an antibody or antibody derivative. This includes chronic and acute disorders or diseases, including pathological conditions that cause mammals to be in trouble.

  A “preservative” is a compound that can be included in a formulation where it can essentially reduce the effects of bacteria, thus facilitating the manufacture of, for example, multi-use formulations. it can. Examples of possible preservatives include octadecyldimethylbenzylammonium chloride, hexamethonium chloride, benzalkonium chloride (a mixture of alkylbenzyldimethylammonium chlorides where the alkyl group is a long chain compound), and benzethonium chloride. It is done. Other types of preservatives include aromatic alcohols such as phenol, butyl and benzyl alcohol, alkyl parabens such as methyl or propyl paraben, catechol, resorcinol, cyclohexanol, 3-pentanol and m-cresol. The most preferred preservative herein is benzyl alcohol.

  The present invention also provides a pharmaceutical composition comprising one or more antibodies or antibody derivative compounds together with at least one physiologically acceptable carrier or excipient. The pharmaceutical composition can be, for example, one or more of water, buffer (eg, neutral buffered saline or phosphate buffered saline), ethanol, mineral oil, vegetable oil, dimethyl sulfoxide, carbohydrate (eg, glucose, mannose). Sucrose or dextran), mannitol, proteins, adjuvants, polypeptides, or amino acids such as glycine, antioxidants, chelating agents such as EDTA or glutathione, and / or preservatives. As mentioned above, other active ingredients may be included in the pharmaceutical compositions provided herein (although not necessary).

  A carrier is a substance that can be associated with an antibody or antibody derivative prior to administration to a patient, often for the purpose of controlling the stability or bioavailability of the compound. Carriers for use within such formulations are generally biocompatible and can also be biodegradable. Carriers include, for example, serum albumin (eg, human or bovine), egg albumin, peptides, monovalent or polyvalent molecules such as polylysine, and polysaccharides such as aminodextran and polyamidoamines. The carrier also includes solid carrier materials such as beads and microparticles including, for example, polylactate, polyglycolate, poly (lactide-co-glycolide), polyacrylate, latex, starch, cellulose or dextran. The carrier can have the compound in a variety of ways, including covalent bonds (either directly or via a linker group), non-covalent interactions or admixtures.

  The pharmaceutical composition can be formulated for any suitable mode of administration, including, for example, topical, oral, nasal, rectal or parenteral administration. In certain embodiments, a composition in a form suitable for oral use is preferred. Such forms include, for example, pills, tablets, troches, lozenges, aqueous or oily suspensions, dispersible powders or granules, emulsions, hard or soft capsules, or syrups or elixirs. Within yet other embodiments, the compositions provided herein can be formulated as a lyophilizate. The term parenteral as used herein refers to subcutaneous, intradermal, intravascular (eg intravenous), intramuscular, spinal, intracranial, intrathecal and intraperitoneal injection, and any similar Including injection or infusion techniques.

  Compositions intended for oral use can be prepared according to any method known in the art for the manufacture of pharmaceutical compositions and include sweeteners to provide attractive and savory preparations. One or more agents such as flavoring agents, coloring agents, and preservatives can be included. Tablets contain the active ingredient in admixture with physiologically acceptable excipients that are suitable for the manufacture of tablets. Such excipients include, for example, inert diluents (eg, calcium carbonate, sodium carbonate, lactose, calcium phosphate, or sodium phosphate), granulating and disintegrating agents (eg, corn starch, or alginic acid), Binders (eg, starch, gelatin, or acacia), and lubricants (eg, magnesium stearate, stearic acid, or talc) are included. The tablets may be uncoated or they may be coated by known techniques to delay disintegration and absorption in the gastrointestinal tract and thereby provide a sustained action over a longer period. For example, a time delay material such as glyceryl monostearate or glyceryl distearate may be employed.

  Formulations for oral use are as hard gelatin capsules in which the active ingredient is mixed with an inert solid diluent (eg, calcium carbonate, calcium phosphate, or kaolin), or the active ingredient is a water or oil vehicle It may be given as a soft gelatin capsule mixed with (eg peanut oil, liquid paraffin or olive oil). Aqueous suspensions contain the antibody or antibody derivative in admixture with excipients suitable for the manufacture of aqueous suspensions. Such excipients include suspending agents (eg, sodium carboxymethylcellulose, methylcellulose, hydropropylmethylcellulose, sodium alginate, polyvinylpyrrolidone, tragacanth gum, and acacia gum), as well as dispersants or humidifiers (eg, , Natural phosphatides such as lecithin, condensation products of alkylene oxides such as polyoxyethylene stearate with fatty acids, condensation products of ethylene oxide such as heptadecaethyleneoxycetanol with long chain aliphatic alcohols, polyoxyethylene Condensation products of polyethylene oxide with partial esters derived from fatty acids such as sorbitol monooleate and hexitol, or polyethylene Condensation products of ethylene oxide with partial esters derived from fatty acids and hexitol anhydrides such as sorbitan monooleate). Aqueous suspensions contain one or more preservatives (eg, ethyl or n-propyl p-hydroxybenzoic acid), one or more colorants, one or more flavoring agents, and one or more. Other sweeteners (eg, sucrose or saccharin). Syrups and elixirs may be formulated with sweetening agents, such as glycerol, propylene glycol, sorbitol, or sucrose. Such formulations can also contain one or more demulcents, preservatives, flavoring agents, and / or coloring agents.

  Oily suspensions may be formulated by suspending the active ingredient in a vegetable oil (eg, arachis oil, olive oil, sesame oil or coconut oil) or in a mineral oil such as liquid paraffin. Oily suspensions may contain a thickening agent, such as beeswax, hard paraffin, or cetyl alcohol. Sweetening agents such as those mentioned above and / or flavoring agents can be added to provide a palatable oral preparation. Such a suspension can be preserved by the addition of an anti-oxidant such as ascorbic acid.

  Dispersible powders and granules suitable for preparation of an aqueous suspension by the addition of water provide the active ingredient in admixture with a dispersing or humidifying agent, suspending agent and one or more preservatives. Suitable dispersing or humidifying agents and suspending agents are exemplified by those already mentioned above. Additional excipients may also be present, for example sweetening, flavoring and coloring agents.

  The pharmaceutical composition may be in the form of an oil-in-water emulsion. The oily phase may be a vegetable oil (eg, olive oil or arachis oil), a mineral oil (eg, liquid paraffin), or a combination of these. Suitable emulsifiers include naturally occurring gums (eg, acacia gum or tragacanth gum), naturally occurring phosphatides (eg, soy, lecithin, and esters or partial esters derived from fatty acids and hexitol), anhydrides (eg, Sorbitan monooleate), and condensation products of fatty acids and partial esters derived from hexitol with ethylene oxide (eg, polyoxyethylene sorbitan monooleate). The emulsion can also contain one or more sweetening and / or flavoring agents.

  Pharmaceutical compositions can be prepared as sterile injectable aqueous or oleaginous suspensions in which the modulator is suspended or dissolved in a vehicle, depending on the vehicle and concentration used. Such compositions can be formulated according to known techniques using suitable dispersing, humidifying and / or suspending agents, such as those described above. Among the acceptable vehicles and solvents that can be employed are water, 1,3-butanediol, Ringer's solution and isotonic saline. In addition, sterile fixed oils can be employed as a solvent or suspending medium. For this purpose any bland fixed oil can be employed including synthetic mono- or diglycerides. In addition, fatty acids such as oleic acid can be used in the preparation of injectable compositions, and adjuvants such as local anesthetics, preservatives and / or buffers can be dissolved in the vehicle.

  The pharmaceutical composition can be formulated as a sustained release formulation (ie, a formulation such as a capsule that provides sustained release of the modulator after administration). Such formulations can generally be prepared using well-known techniques and can be administered, for example, by oral, rectal or subcutaneous implantation, or by implantation at the desired target site. Carriers for use in such formulations are biocompatible and can also be biodegradable, preferably the formulation allows a relatively constant level of modulator release. The amount of antibody or antibody derivative contained within the sustained release formulation will depend, for example, on the site of implantation, the rate of release and the expected duration, and the characteristics of the disease / disorder to be treated or prevented.

  The antibodies or antibody derivatives provided herein generally bind to a target, such as VEGF, to a detectable extent and prevent or inhibit a target-mediated disease / disorder, such as a VEGF-mediated disease / disorder. Administer in an amount that reaches a concentration in the body fluid (eg, blood, plasma, serum, CSF, synovial fluid, lymph, interstitial fluid, tears or urine) sufficient to do so. A dose is considered effective when it provides a discernible patient benefit as described herein. Preferred systemic doses range from about 0.1 mg to about 140 mg per kilogram body weight per day (about 0.5 mg to about 7 g per patient per day), and oral doses are generally About 5 to 20 times more than the intravenous dose. The amount of antibody or antibody derivative that can be combined with a carrier material to produce a single dosage form will vary depending upon the host treated and the particular mode of administration. Dosage unit forms will generally contain between from about 1 mg to about 500 mg of an active ingredient.

  The pharmaceutical composition can be packaged, for example, to treat a condition responsive to an antibody or antibody derivative directed against VEGF. The packaged pharmaceutical composition comprises a container holding an effective amount of at least one antibody or antibody derivative as described herein, and a disease / disorder responsive to one antibody or antibody derivative after administration to a patient Instructions (e.g., labels) can be included that indicate the use of the containing composition to treat.

The antibody or antibody derivative of the present invention can also be chemically modified. Preferred modifying groups are polymers such as, for example, optionally substituted linear or branched polyalkenes, polyalkenylene or polyoxyalkylene polymers, or polysaccharides that are branched or unbranched. Such effector groups can increase the half-life of the antibody in vivo. Specific examples of synthetic polymers include optionally substituted linear or branched poly (ethylene glycol) (PEG), poly (propylene glycol), poly (vinyl alcohol) or derivatives thereof. Specific polymers that occur in nature include lactose, amylose, dextran, glycogen or derivatives thereof. The size of the polymer can vary as desired, but typically the average molecular weight is in the range of 500 Da to 50000 Da. If the antibody is designed to penetrate tissue for local application, the preferred molecular weight of the polymer is about 5000 Da. The polymer molecule can be attached to the C-terminal part of the heavy chain of the Fab fragment via an antibody, in particular a hinge peptide covalently linked, as described in WO0194585. Regarding the attachment of the PEG moiety, “Poly (ethyleneglycol) Chemistry,
Biotechnological and Biomedical Applications ", 1992, J. MoI. Milton Harris (eds.), Plenum Press, New York, and “Bioconjugation Protein Coupling Technologies for the Biomedical
Sciences ", 1998, M.C. Aslam and A.M. See Dent, Globe Publishers, New York.

  After the target antibody or antibody derivative is prepared as described above, a pharmaceutical formulation containing it is prepared. The formulated antibody is not subjected to prior lyophilization, and the formulation of interest herein is an aqueous formulation. Preferably, the antibody or antibody derivative in the formulation is an antibody fragment such as scFv. The therapeutically effective amount of antibody present in the formulation is determined, for example, taking into account the desired dose volume and mode (s) of administration. From about 0.1 mg / ml to about 50 mg / ml, preferably from about 0.5 mg / ml to about 25 mg / ml, and most preferably from about 2 mg / ml to about 10 mg / ml are typical antibody concentrations in the formulation. It is.

  Aqueous formulations are prepared containing the antibody or antibody derivative in a pH buffered solution. The buffer of the present invention has a pH in the range of about 4.5 to about 6.0, preferably about 4.8 to about 5.5, and most preferably about 5.0. Examples of buffers that control pH within this range include acetate (eg, sodium acetate), succinate (eg, sodium succinate), gluconate, histidine, citrate and other organic acid buffers Is mentioned. The concentration of the buffer can be from about 1 mM to about 50 mM, preferably from about 5 mM to about 30 mM, and can be determined, for example, by the desired isotonicity of the buffer and the formulation. A preferred buffer is sodium acetate (about 10 mM), pH 5.0.

  A polyol that can act as a tonicifier and stabilize the antibody is included in the formulation. In a preferred embodiment, the formulation does not include a salt such as sodium chloride in a tonicifying amount because it may cause precipitation of the antibody or antibody derivative and / or cause oxidation at low pH. In a preferred embodiment, the polyol is a non-reducing sugar such as sucrose or trehalose. The polyol is added to the formulation in an amount that can vary for the desired isotonicity of the formulation. Preferably, the aqueous formulation is isotonic, in which case a suitable polyol concentration in the formulation is, for example, in the range of about 1% to about 15% w / v, preferably about 2% to about 10% whv. Range. However, hypertonic or hypotonic formulations may also be appropriate. The amount of polyol added can also vary with respect to the molecular weight of the polyol. For example, a smaller amount of monosaccharide (eg, mannitol) can be added compared to a disaccharide (such as trehalose).

  Surfactants are also added to the antibody or antibody derivative formulation. Exemplary surfactants include nonionic surfactants such as polysorbates (eg, polysorbate 20, 80, etc.) or poloxamers (eg, poloxamer 188). The amount of surfactant added is such that it reduces aggregation of the formulated antibody / antibody derivative and / or minimizes the formation of particles in the formulation and / or reduces adsorption. For example, the surfactant is in an amount of about 0.001% to about 0.5%, preferably about 0.005% to about 0.2%, and most preferably about 0.01% to about 0.1%. Can be present in the formulation.

  In one embodiment, the formulation comprises the agent identified above (ie, antibody or antibody derivative, buffer, polyol and surfactant) and consists essentially of benzyl alcohol, phenol, m-cresol, chlorobutanol and It does not contain one or more preservatives such as benzethonium chloride. In another embodiment, preservatives can be included in the formulation, particularly when the formulation is a multi-dose formulation. The concentration of preservative may range from about 0.1% to about 2%, most preferably from about 0.5% to about 1%. Remington's Pharmaceutical Sciences 21st edition, Osol, A.M. One or more other pharmaceutically acceptable carriers, excipients or stabilizers, such as those described in Ed. (2006), provided they do not adversely affect the desired characteristics of the formulation And can be included in the formulation. Acceptable carriers, excipients or stabilizers are not toxic to the recipient at the dosage and concentration employed, and include additional buffers, co-solvents, antioxidants including ascorbic acid and methionine, chelating agents such as EDTA , Metal complexes (eg Zn-protein complexes), biodegradable polymers such as polyester, and / or salt-forming counterions such as sodium.

  Formulations used for in vivo administration must be sterile. This is easily accomplished by filtration through sterile filtration membranes before or after preparing the formulation.

  The formulation may be administered to a mammal, preferably a human, in need of treatment with an antibody, intramuscular, intraperitoneal, intracerebral spinal, subcutaneous, intraarticular, intrasynovial, intrathecal, oral, topical. Or it is administered according to known methods such as intravenous administration by bolus or by continuous infusion over a period of time, by inhalation route. In a preferred embodiment, the formulation is administered to the mammal by intravenous administration. For such purposes, the formulation can be injected, for example, using a syringe or via the IV line.

  Appropriate doses of antibody ("therapeutically effective amount") can be determined, for example, by the condition being treated, the severity and course of the condition, whether the antibody is administered for prophylactic or therapeutic purposes, previous therapy, patient history It depends on the response to the antibody, the type of antibody used and the discretion of the attending physician. The antibody or antibody derivative can be suitably administered to the patient at one time or over a series of treatments and administered to the patient at any time after diagnosis. The antibody or antibody derivative can be administered as a single treatment or in conjunction with other drugs or therapies useful in the treatment of the condition in question.

  As a general proposition, the therapeutically effective amount of antibody or antibody derivative to be administered, whether in a single dose or multiple doses, will range from about 0.1 to about 50 mg / kg of the patient's body weight. The typical range of antibodies used will be, for example, from about 0.3 to about 20 mg / kg, more preferably from about 0.3 to about 15 mg / kg for daily administration. However, other dosing regimens may be useful. The progress of this therapy is easily monitored by conventional techniques.

  In another embodiment of the invention, there is provided an article of manufacture comprising a pharmaceutical formulation of the invention, preferably a container holding an aqueous formulation, and optionally instructions for its use. Suitable containers include, for example, bottles, vials and syringes. The container can be formed from a variety of materials such as glass or plastic. A typical container is a 3-20 cc single use glass vial. Alternatively, for multi-dose formulations, the container may be a 3-100 cc glass vial. The container holds the formulation and can indicate directions for use by a label on the container or by a label associated with the container. The article of manufacture can further include other materials desired from a commercial and user perspective, including other buffers, diluents, filters, needles, syringes, and package inserts with instructions for use. Is mentioned.

  In certain preferred embodiments, the article of manufacture comprises a lyophilized immunobinder described herein, or a lyophilized immunobinder produced by the methods described herein.

(Example)
The present disclosure is further illustrated by the following examples that should not be construed as further limiting. The contents of all figures and all references, patents and published patent applications cited throughout this application are expressly incorporated herein by reference in their entirety.

(Materials and methods)
In general, practice of the invention employs conventional techniques of chemistry, molecular biology, recombinant DNA techniques, immunology (especially antibody techniques, for example), and standard techniques for polypeptide preparation, unless otherwise indicated. See, for example, Sambrook, Fritsch and Maniatis, Molecular Cloning: Cold Spring Harbor Laboratory Press (1989); Antibody Engineering Protocols (Methods in Molecular S), Methods in LoS. , Humana Pr (1996); Antibody Engineering: A Practical Approach (Practical Approaches Series, 169), McCafety ed., Irl Pr (1996); S. H. L. Press, Pub. (1999); and Current Protocols in Molecular Biology, edited by Ausubel et al., John Wiley & Sons (1992). Methods for fusion of CDRs from rabbits and other non-human monoclonal antibodies to selected human antibody frameworks are described in detail above. Examples of such fusion experiments are shown below.

  For better understanding, a fusion fragment named “min” is one in which the CDR is fused to framework 1.4 or its variable domain, while a fusion fragment named “max” , The CDR is fused to framework 1.4 or a variable domain thereof, and the framework further comprises donor framework residues that interact with the antigen.

(Example 1: Design of rFW1.4)
(1.1. Primary sequence analysis and database search)
(1.1.1. Collection of rabbit immunoglobulin sequences)
The variable domain and germline sequences of mature rabbit antibodies were collected from various open source databases (eg, Kabat database and IMGT) and entered into the customized database as single-letter encoded amino acid sequences. For all analyses, we used only the amino acid portion corresponding to the V (variable) region. Sequences in the KDB database that contained less than 70% completeness or multiple undecided residues within the framework regions were discarded. Sequences with greater than 95% identity to any other sequence in the database were also excluded to avoid random noise in the analysis.

(1.1.2. Alignment and numbering of rabbit sequences)
The rabbit antibody sequences were aligned using a conventional sequence alignment tool based on the Needleman-Wunsch algorithm and Blossum matrix. The introduction of gaps and nomenclature of residue positions was performed according to the AHo numbering system of immunoglobulin variable domains (Honegger and Puckthun, 2001). The Kabat numbering scheme was also used because it is the most widely adopted standard method for numbering residues in antibodies. Kabat numbering was assigned using the SUBIM program. This program analyzes the variable regions of antibody sequences and numbers the sequences according to the system established by Kabat et al. (Deret et al., 1995).

  The framework and CDR regions are defined according to the Kabat definition, which is based on sequence variability and is most commonly used. Nevertheless, the designation CDR-H1 is the definition of AbM, Kabat, average contact data generated by analysis of contacts between antibodies and antigens of a subset of 3D complex structures (MacCallum et al., 1996). ) And various definitions including Chotia's definition based on the location of the structural loop region (as shown above and shown in FIG. 1).

(1.1.3. Frequency and conservation of residue positions)
Amino acid sequence diversity was analyzed using a set of 423 rabbit sequences obtained from the kabat database. The residue frequency f (r) at each position i in the mature rabbit sequence was calculated by dividing the number of times a particular residue was observed in the data set by the total number of sequences. The degree of conservation at each position i was calculated using the Simpson index, taking into account the number of different amino acids present and the relative abundance of each residue.

Where N is the total number of amino acids, r is the number of different amino acids present at each position, and n is the number of residues of a particular amino acid type.

(1.1.4. Series analysis of rabbit V region)
The rabbit repertoire was investigated using phylogenetic analysis tools. V region amino acid sequences were clustered using both cluster and topological algorithms. A distance matrix was calculated for the entire array and used as an indicator of germline usage. The consensus sequence for each cluster was calculated and the closest rabbit germline sequence counterpart was identified. A comprehensive consensus sequence was also obtained from the entire set of sequences.

(1.1.5. Assignment of human subgroups)
Identify the most homologous human subgroup for each rabbit representative sequence in various clusters using a taxonomy based on the EXCEL implementation of the sequence analysis algorithm and analysis of the human antibody repertoire (Knappik et al., 2000) did.

(1.2. Design of human acceptor framework)
The rabbit repertoire blueprint from the sequence analysis described above was used to identify residues in the framework that are commonly involved in positioning the rabbit CDRs. Of the frameworks with high homology to the rabbit repertoire and each cluster, one with good biophysical properties was selected from a pool of fully human sequences. The frameworks selected for use as acceptor frameworks belong to variable light chain subgroup kappa 1 and variable heavy chain subgroup III, corresponding to ESBATtech ID KI27, a43. This stable and soluble antibody framework was identified by screening a human spleen scFv library using a yeast-based screening method (Auf der Maur et al., 2004) termed a “quality control” system. It was named “FW1.4”. Although the stable and soluble framework sequence FW1.4 shows high homology, it was not the highest homologous sequence available. The identified residues were incorporated into the acceptor framework to generate rFW1.4.

Using information about the amino acid sequence diversity, germline usage, and structural characteristics of rabbit antibodies, we can match the residue positions necessary to conserve CDR conformation in the novel human framework. FW1.4 was analyzed for. We examined the variable region of FW1.4 for suitability for the following features:
i. Residues that are part of the reference sequence of the loop structure ii. Framework residues located at the VL / VH interface iii. Residue platform just below the CDR iv. Upper and lower core residues v. Framework residues that define subtypes.

(1.3. Fusion of rabbit CDR)
The fusion piece was generated by simply combining the CDR sequence from one antibody (as defined above) with the framework sequence of FW1.4 or rFW1.4. Residues potentially involved in binding were identified. For each rabbit variable domain sequence, the closest rabbit germline counterpart was identified. If the closest germline could not be established, the sequences were compared against a subgroup consensus or a consensus of rabbit sequences with a high percent similarity. Rare framework residues could be the result of somatic hypermutation and were therefore considered to play a role in binding. Therefore, such residues were fused to the acceptor framework.

(1.4. Results)
By analyzing the rabbit antibody repertoire in terms of structure, amino acid sequence diversity and germline use, five residue positions in the light chain of FW1.4 were found that were modified to maintain the loop conformation of the rabbit CDR. It was. These positions are highly conserved with rabbit antibodies. These five consensus residues were deduced from the rabbit repertoire and introduced into the human acceptor framework 1.4. These conserved site modifications made the framework compatible with virtually all six complementarity determining regions (CDRs) of any rabbit CDR. Master rFW1.4 containing various rabbit CDRs was well expressed and produced well as opposed to wild type single chain. Sixteen members resulting from the combination of this framework and the rabbit CDRs were generated and detailed characterization demonstrated functionality.

(Example 2: B cell screening system)
ESBATtech has established a FACS (flow cytometry-based single cell sorting) -based screening system for selecting B cells that bind to their target of interest through their B cell receptor (BCR). One target was, for example, a soluble protein, a single chain antibody (ESBA903) labeled with fluorescent dyes (PE and PerCP). Lymphocyte suspensions were prepared from the spleens of rabbits immunized with recombinant targets. The cells were then incubated with PE and PerCP labeled ESBA903 and IgG specific antibody (APC label) or IgM specific antibody (FITC label). ESBA903 positive B cells expressing IgG on their surface but not IgM were sorted into 96 well plates (FIG. 3; Table 2). B which has grown and differentiated into plasma cells and secreted antibodies by a thymoma helper cell line (EL4-B5: Zubler et al., 1985, J. Immunol, 134 (6), 3662-3668). Cells were selected. The affinity of these IgGs for the target was assessed by ELISA and Biacore measurements. The kinetic parameters of 7 selected clones are shown in Table 1. These clones, obtained from a pool of approximately 200 sorted cells, exhibit binding affinities in the low nanomolar to picomolar range. Finally, mRNA was isolated from 6 clones of interest and CDRs were fused to the single chain framework FW1.4.

(Example 3: Detection of interaction between anti-TNFα antibody-coated beads and CHO cells expressing membrane-bound TNFα)
To evaluate whether high pressure in the flow cytometry stream breaks the non-covalent bond between the two cells, the following experiment was performed. CHO cells stably transfected with membrane-bound TNFα (B-220 cells) were incubated with beads coated with PE-labeled anti-TNFα antibody. In this setup, the beads mimic memory B cells (they have approximately the same size). As negative controls, non-transfected CHO cells and beads coated with an APC-labeled irrelevant antibody (anti-CD19) were used. After 2 hours incubation at 4 ° C. with agitation, the cell bead suspension was analyzed by FACS (using a 130 μm nozzle). FIG. 4 shows that specific binding between anti-TNFα beads and TNFα-transfected CHO cells is clearly detectable with FACS. In fact, in this sample (upper panel), about 2/3 of the beads are bound to the cells (585 is bound, whereas 267 is unbound). In contrast, in the control samples (middle and lower panels), most beads do not bind to CHO cells. In addition, both bead populations (anti-TNFα-PE and anti-CD19-APC) were mixed with TNFα-transfected CHO cells. FIG. 5 and Table 4 show that about half of the anti-TNFα beads bind to CHO cells, while the majority of the anti-CD19 beads remain unbound. The percentage of cell-bound beads in each sample is detailed in Table 5. Thus, we demonstrate that specific selection of single B cells that bind to membrane-bound target proteins via their B cell receptors is possible using flow cytometry.

Example 4: CDR fusion and functional humanization of anti-TNFα rabbit donor antibody
Four anti-TNFα rabbit antibodies “Rabmab” (EPI-1, EPI-15, EP-34, EP-35 and EP-42) were selected for CDR fusion. The general experimental scheme for CDR fusion, humanization, and preliminary characterization of the humanized rabbit donor antibody was performed as outlined in the detailed description. Unlike traditional humanization methods that utilize human antibody acceptor frameworks that share the greatest sequence homology with non-human donor antibodies, quality control assays (WO0148017) are used to achieve the desired functional properties (solubility and stability). The rabbit CDRs were fused to a human framework (FW 1.4) preselected for gender. These stable and soluble framework sequences showed high homology to RabMab.

  Using the methodology described herein, CDR fusion fragments were generated for each of the RabMabs. In the “Min” fusion, only the rabbit CDRs were grafted from the VL and VH domains of the rabbit donor antibody to the human acceptor framework FW1.4. The “Max” fusion piece refers to the fusion of a rabbit CDR to rFW1.4.

The scFv described and characterized herein was produced as follows. The humanized VL sequences via a linker of SEQ ID NO: 8 are connected to the humanized VH _sequences, resulted in scFv orientation of NH 2 -VL- linker -VH-COOH. In many cases, DNA sequences encoding various scFvs were newly synthesized by the service provider Enterechon GmbH (www.enterechon.com). The resulting DNA insert was cloned into the bacterial expression vector pGMP002 via NcoI and HindIII restriction enzyme recognition sites introduced at the 5 ′ and 3 ′ ends of the scFv DNA sequence, respectively. A BamHI restriction enzyme recognition site is located between the DNA sequence of the VL domain and the DNA sequence of the VH domain. In some cases, scFv encoding DNA was not synthesized de novo and the scFv expression constructs were cloned by domain shuffling. Therefore, the VL domain was excised and introduced into the new construct via the NcoI and BamHI restriction enzyme recognition sites, and the VH domain was excised and introduced via the BamHI and HindIII restriction enzyme recognition sites. In other cases, point mutations were introduced into the VH and / or VL domains using state-of-the-art assembly PCR methods. Cloning of GMP002 is described in Example 1 of WO2008006235. The production of scFv was performed in the same manner with respect to ESBA105 described in Example 1 of WO2008006235.

  Table 3 summarizes the detailed characterization data for the four rabbit monoclonals (EP6, EP19, EP34, EP35 and EP43) and their CDR fusion variants. These CDR fusions showed extensive activity in the BIACore binding assay and the L929, TNFα-mediated cytotoxicity assay, but 3 of the 4 largest (“max”) fusions showed therapeutically significant activity. It was. EP43max showed the most favorable binding affinity (0.25 nM Kd) and excellent EC50 in cytotoxicity assays. This data indicates that FW1.4 (SEQ ID NOs: 1 and 2) is an exemplary soluble and stable human acceptor framework region for rabbit CDR fusions.

(Efficacy assay)
The neutralizing activity of anti-TNFα binders was assessed with the L929 TNFα-mediated cytotoxicity assay. Toxicity of mouse L929 fibroblasts treated with actinomycin was induced with recombinant human TNF (hTNF). 90% of maximal hTNF-induced cytotoxicity was determined to be a TNF concentration of 1000 pg / ml. All L929 cells were cultured in RPMI 1640 containing phenol red in L-glutamine medium supplemented with fetal calf serum (10% v / v). The neutralizing activity of the anti-TNFα binder was evaluated in RPMI 1640 without phenol red and 5% fetal calf serum. To determine the concentration at which antagonism reaches half-maximal inhibition (EC 50%), various concentrations (0-374 ng / mL) of anti-TNF binder are added to L929 cells in the presence of 1000 pg / ml hTNF. A dose response curve was fitted to a non-linear sigmoidal regression with variable slope and an EC50 was calculated.

(Biacore binding analysis of anti-TNF scFv)
For binding affinity measurement, surface plasmon resonance measurement using BIAcore ^™ -T100 was utilized using NTA sensor chip and His-tagged TNF (generated at ESBATtech). The surface of the NTA sensor chip consists of a carboxymethylated dextran matrix pre-immobilized with nitrilotriacetic acid (NTA) to capture histidine-tagged molecules via Ni2 + NTA chelation. Human TNFα N-his trimer (5 nM) was captured by nickel via their N-terminal his tag and ESBA105 (analyzed at several concentrations ranging from 30 nM to 0.014 nM in a 3-fold serial dilution step. Stuff). In the regeneration step, the complexes formed by the nickel, ligand and analyte are washed away. This allows the same regeneration conditions to be used for different samples. Reaction signals are generated by surface plasmon resonance (SPR) technique and measured in resonance units (RU). All measurements are performed at 25 ° C. A sensorgram was generated for each anti-TNF scFv sample after inline reference cell correction and subsequent subtraction of buffer samples. Apparent dissociation rate constant (k _d ), apparent association rate constant (k _a ), and apparent dissociation equilibrium constant (K _D ), one-to-one Langmuir binding using BIAcore T100 evaluation software version 1.1 Calculated using the model.

Example 5: CDR fusion and functional humanization of anti-VEGF rabbit donor antibody
Eight anti-VEGF Rabmabs (375, 435, 509, 511, 534, 567, 578 and 610) were selected for CDR fusion. Unlike traditional humanization methods that utilize human antibody acceptor frameworks that share the greatest sequence homology with non-human donor antibodies, quality control assays (WO0148017) are used to achieve the desired functional properties (solubility and stability). The rabbit CDRs were fused to the human acceptor framework FW1.4 (SEQ ID NOs: 1 and 2) preselected for gender.

  Using the methodology described herein, a number of CDR fusion fragments were generated for each of RabMab (the rabbit antibody) (see Example 4). The “Min” fusion piece contained the smallest fusion piece. In this, only the rabbit CDRs were transplanted from the VL and VH domains of the rabbit donor antibody into the human acceptor framework FW1.4 (SEQ ID NO: 1). The “Max” fusion fragment included not only the rabbit CDRs for VL and VH, but also some additional framework residues from rabbit donors that were predicted to be important for antigen binding. For 578max, the heavy chain variable domain framework region of FW1.4 has additional amino acid modifications at Kabat positions 23H, 49H, 73H, 78H and 94H.

  Table 4 summarizes the detailed characterization data for the “Min” and “Max” CDR fusion variants. FIG. 5 shows their efficacy as VEGF inhibitors as measured using VEGFR competition ELISA and / or HUVEC assays. This data indicates that FW1.4 (SEQ ID NOs: 1 and 2) is an exemplary soluble and stable human acceptor framework region for rabbit CDR fusions.

(Biacore binding analysis of anti-VEGF scFv)
The Biacore binding ability of the scFv was tested and binding affinity was measured using an exemplary surface plasmon resonance method with BIAcore ^™ -T100. In this and subsequent examples, VEGF proteins tested for binding by these scFv candidates include purified Escherichia coli-expressed recombinant human VEGF ₁₆₅ (PeproTech EC Ltd.), recombinant human VEGF ₁₂₁ (PeproTech EC). Ltd.), recombinant human VEGF ₁₁₀ (ESBATtech AG), recombinant mouse VEGF ₁₆₄ (PeproTech EC Ltd.), recombinant rat VEGF ₁₆₄ (Biovision), recombinant rabbit VEGF ₁₁₀ (ESBATtech AG), and recombinant human PLGF. (PeproTech EC Ltd.). For surface plasmon resonance experiments, a carboxymethylated dextran biosensor chip (CM4, GE Healthcare) was prepared using N-ethyl-N ′-(3-dimethylaminopropyl) carbodiimide hydrochloride and N according to the supplier's instructions. -Activated with hydroxysuccinimide. Each of the six different VEGF forms exemplified above were bound to one of four different flow cells on a CM4 sensor chip using standard amine coupling procedures. After coupling and blocking, the range of responses obtained with these immobilized VEGF molecules is approximately 250-500 response units (RU) for hVEGF ₁₆₅ , hVEGF ₁₁₀ , hVEGF ₁₂₁ , mouse VEGF ₁₆₄ , rat VEGF ₁₆₄ , and It was about 200 RU for rabbit VEGF ₁₁₀ and about 400 RU for PLGF. The fourth flow cell of each chip was treated similarly, except that the protein was not immobilized prior to blocking, and this flow cell was used as an inline reference. Various concentrations of anti-VEGF in HBS-EP buffer (0.01 M HEPES, pH 7.4, 0.15 M NaCl, 3 mM EDTA, 0.005% surfactant P20)
scFv (eg, 90 nM, 30 nM, 10 nM, 3.33 nM, 1.11 nM, 0.37 nM, 0.12 nM and 0.04 nM) was injected into the flow cell for 5 minutes at a flow rate of 30 μl / min. Dissociation of anti-VEGF scFv from VEGF on the CM4 chip was allowed to proceed for 10 minutes at 25 ° C. After subtracting buffer samples following inline reference cell correction, a sensorgram was generated for each anti-VEGF scFv sample. Apparent dissociation rate constant (k _d ), apparent association rate constant (k _a ), and apparent dissociation equilibrium constant (K _D ), using the BIAcoreT100 evaluation software version 1.1, a one-to-one Langmuir binding model Calculated using

(HUVEC assay of VEGF inhibition)
The HUVEC assay is a method for measuring the efficacy of the disclosed anti-VEGF scFv candidates as a VEGF inhibitor.

Human umbilical vein endothelial cells (HUVEC) (Promocell) pooled from several donors were used from passage 2 to passage 14. Cells were treated with 0.4% ECGS / H, 2% fetal calf serum, 0.1 ng / ml epidermal growth factor, 1 μg / ml hydrocortisone, 1 ng / ml basic fibroblast factor and 1% penicillin / streptomycin. Seeded at 1000 cells / well in 50 μl complete endothelial cell growth medium (ECGM) (Promocell) containing (Gibco). After 7 to 8 hours, 50 μl of starvation medium (ECGM without supplement containing 0.5% heat inactivated FCS and 1% penicillin / streptomycin) was added to the cells and the cells were allowed to starve for 15 to 16 hours. It was. Three-fold serial dilutions of anti-VEGF scFv (0.023-150 nM) and recombinant human VEGF ₁₆₅ (0.08 nM), recombinant mouse VEGF ₁₆₄ (0.08 nM) or recombinant rat VEGF ₁₆₄ (0.3 nM) One was prepared in starvation medium and preincubated for 30-60 minutes at room temperature. Different concentrations of VEGF were used to compensate for their different relative biological activities. A concentration that stimulated submaximal VEGF-induced proliferation (EC ₉₀ ) was used. 100 μl of the mixture was added to a 96-well tissue culture plate containing HUVEC suspension and incubated for 4 days in a humidified incubator at 37 ° C./5% CO ₂ . Using a Sunrise microplate reader (Tecan), add 20 μl / well WST-1 cell growth reagent (Roche), then measure the proliferation of HUVEC at 450 nm (using 620 nm as reference wavelength). Was evaluated by Data was analyzed using a four parameter logistic curve fit and the concentration of anti-VEGF scFv required for 50% HUVEC growth inhibition (EC ₅₀ ) was derived from the inhibition curve.

(Equivalent)
Many modifications and alternative embodiments of the invention will be apparent to those skilled in the art in view of the foregoing description. Accordingly, this description is to be construed as illustrative only and is for the purpose of teaching those skilled in the art the best mode of carrying out the invention. The details of construction may vary substantially without departing from the spirit of the invention and retain exclusive use rights for all modifications that fall within the scope of the appended claims. It is intended that the present invention be limited only to the extent required by the appended claims and the rules of applicable law.

  All documents and similar materials cited in this application, including patents, patent applications, editorials, books, papers, articles, web pages, figures and / or attachments, are referenced, regardless of the format of such documents and analogs. Are expressly incorporated by reference in their entirety. In the event that one or more of the incorporated literature and analogs differs from or contradicts this specification, including defined terms, term usage, described techniques, or the like, this specification have priority.

Claims (19)

(I) light chain CDR1, CDR2 and CDR3 from the donor rabbit antibody or antigen-binding fragment thereof and heavy chain CDR1, CDR2 and CDR3 from the donor rabbit antibody or antigen-binding fragment thereof;
(Ii) a human variable light chain framework; and (iii) the following amino acids: threonine (T) at position 24, alanine (A) or glycine (G) at position 56, threonine (T) at position 84, at position 89 comprising human variable heavy chain framework comprising at least four of the valine (V) and position 108 arginine (R) (AHo numbering), an antibody or antigen-binding fragment thereof, of the variable heavy chain framework An antibody or antigen-binding fragment thereof, wherein the amino acid sequence is at least 90% identical to the sequence of SEQ ID NO: 4 . The variable heavy chain framework is
(A) Threonine (T) at position 24 (AHo numbering);
(B) Threonine (T) at position 84 (AHo numbering); or (c) Valine (V) at position 89 (AHo numbering).
The antibody or antigen-binding fragment thereof according to claim 1, comprising:   The variable heavy chain framework comprises the following amino acids: serine (S) at position 12, serine (S) or threonine (T) at position 103, and serine (S) or threonine (T) at position 144 (AHo numbering). The antibody or antigen-binding fragment thereof of claim 1, further comprising at least one of:   2. The antibody or antigen-binding fragment thereof of claim 1, wherein the variable heavy chain framework further comprises glycine (G) at position 141 (AHo numbering).   The antibody or antigen-binding fragment thereof according to claim 1, wherein the variable heavy chain framework comprises an amino acid sequence selected from the group consisting of SEQ ID NO: 4 and SEQ ID NO: 6.   2. The antibody or antigen binding fragment thereof of claim 1, wherein the variable light chain framework comprises an amino acid sequence that is at least 90% identical to the sequence of SEQ ID NO: 2. The antibody or antigen-binding fragment thereof of claim 6 , wherein the variable light chain framework comprises SEQ ID NO: 2. 7. The antibody or antigen-binding fragment thereof of claim 6 , wherein the variable light chain framework comprises threonine (T) at position 87 (AHo numbering). 9. The antibody or antigen-binding fragment thereof of claim 8 , wherein the variable light chain framework comprises SEQ ID NO: 9. The antibody or antigen-binding fragment thereof of claim 6 , wherein the variable heavy chain framework and the variable light chain framework are linked by a linker sequence comprising SEQ ID NO: 8. The antibody or antigen-binding fragment thereof according to claim 6 , comprising an amino acid sequence that is at least 90% identical to the sequence of SEQ ID NO: 3, SEQ ID NO: 5 or SEQ ID NO: 7. SEQ ID NO: 3, comprising an amino acid sequence selected from the group consisting of SEQ ID NO: 5 or SEQ ID NO: 7, an antibody or antigen-binding fragment thereof of claim 1 1. The variable light chain framework is the following amino acids: glutamic acid at position 1 (E), valine at position 3 (V), leucine at position 4 (L), serine at position 10 (S), arginine at position 47 (R) The antibody or antigen-binding fragment thereof according to claim 6 , further comprising at least one of: serine (S) at position 57, phenylalanine (F) at position 91 and valine (V) at position 103 (AHo numbering). . The antibody or antigen-binding fragment thereof according to claim 6 , wherein the antibody or antigen-binding fragment thereof is scFv.   A pharmaceutical composition comprising the antibody or antigen-binding fragment thereof according to claim 1 and a pharmaceutically acceptable excipient.   2. The antibody or antigen-binding fragment thereof of claim 1, wherein the human variable heavy chain framework further comprises valine (V) at position 25 and / or lysine (K) at position 82 (AHo numbering). Bispecific molecule comprising an antibody or antigen-binding fragment thereof of any one of claims 1 to 1 4 or 1 6. The bispecific molecule of claim 17 that binds to at least two different binding sites or target molecules. The antibody or antigen-binding fragment thereof is an antibody, antibody fragment (including Fab, Fab ′, F (ab ′) ₂ , Fv, or single chain Fv), tumor-specific or pathogen-specific antigen, peptide, or binding The bispecific molecule of claim 17 or 18 linked to a sexual mimetic.